{
  "best_metric": 0.9642074088296352,
  "best_model_checkpoint": "checkpoints/BEE-spoke-data-bert-plus-L8-v1.0-allNLI_matryoshka-synthetic-text-similarity-Mar-07_22-56/checkpoint-10800",
  "epoch": 0.8683417085427135,
  "eval_steps": 300,
  "global_step": 10800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 7.8406219482421875,
      "learning_rate": 8.038585209003216e-08,
      "loss": 0.1372,
      "step": 5
    },
    {
      "epoch": 0.0,
      "grad_norm": 6.115014553070068,
      "learning_rate": 1.6077170418006432e-07,
      "loss": 0.1134,
      "step": 10
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.3239176273345947,
      "learning_rate": 2.4115755627009647e-07,
      "loss": 0.0902,
      "step": 15
    },
    {
      "epoch": 0.0,
      "grad_norm": 4.643126964569092,
      "learning_rate": 3.2154340836012864e-07,
      "loss": 0.0907,
      "step": 20
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.8138389587402344,
      "learning_rate": 4.019292604501608e-07,
      "loss": 0.087,
      "step": 25
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.0662753582000732,
      "learning_rate": 4.823151125401929e-07,
      "loss": 0.0713,
      "step": 30
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.2391765117645264,
      "learning_rate": 5.627009646302252e-07,
      "loss": 0.0724,
      "step": 35
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.7336483001708984,
      "learning_rate": 6.430868167202573e-07,
      "loss": 0.0711,
      "step": 40
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.5051072239875793,
      "learning_rate": 7.234726688102894e-07,
      "loss": 0.0386,
      "step": 45
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.8692330121994019,
      "learning_rate": 8.038585209003216e-07,
      "loss": 0.07,
      "step": 50
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.106194257736206,
      "learning_rate": 8.842443729903538e-07,
      "loss": 0.0908,
      "step": 55
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.4909865856170654,
      "learning_rate": 9.646302250803859e-07,
      "loss": 0.0632,
      "step": 60
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.8705946207046509,
      "learning_rate": 1.045016077170418e-06,
      "loss": 0.0436,
      "step": 65
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.9496204853057861,
      "learning_rate": 1.1254019292604503e-06,
      "loss": 0.0481,
      "step": 70
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.2997841835021973,
      "learning_rate": 1.2057877813504825e-06,
      "loss": 0.0552,
      "step": 75
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.0006277561187744,
      "learning_rate": 1.2861736334405146e-06,
      "loss": 0.0249,
      "step": 80
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.1965364217758179,
      "learning_rate": 1.3665594855305468e-06,
      "loss": 0.0261,
      "step": 85
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.8660438656806946,
      "learning_rate": 1.4469453376205788e-06,
      "loss": 0.0383,
      "step": 90
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.8808113932609558,
      "learning_rate": 1.527331189710611e-06,
      "loss": 0.0254,
      "step": 95
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.918907880783081,
      "learning_rate": 1.6077170418006432e-06,
      "loss": 0.0656,
      "step": 100
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.735292673110962,
      "learning_rate": 1.6881028938906755e-06,
      "loss": 0.0518,
      "step": 105
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.8118743896484375,
      "learning_rate": 1.7684887459807077e-06,
      "loss": 0.0908,
      "step": 110
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.0685555934906006,
      "learning_rate": 1.84887459807074e-06,
      "loss": 0.0332,
      "step": 115
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.8618605732917786,
      "learning_rate": 1.9292604501607717e-06,
      "loss": 0.0456,
      "step": 120
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.7448190450668335,
      "learning_rate": 2.0096463022508037e-06,
      "loss": 0.0349,
      "step": 125
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6762810945510864,
      "learning_rate": 2.090032154340836e-06,
      "loss": 0.025,
      "step": 130
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.172909736633301,
      "learning_rate": 2.170418006430868e-06,
      "loss": 0.0334,
      "step": 135
    },
    {
      "epoch": 0.01,
      "grad_norm": 5.036996364593506,
      "learning_rate": 2.2508038585209006e-06,
      "loss": 0.0548,
      "step": 140
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.498236894607544,
      "learning_rate": 2.3311897106109326e-06,
      "loss": 0.0492,
      "step": 145
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.8583147525787354,
      "learning_rate": 2.411575562700965e-06,
      "loss": 0.05,
      "step": 150
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.241025686264038,
      "learning_rate": 2.491961414790997e-06,
      "loss": 0.0264,
      "step": 155
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.9725112915039062,
      "learning_rate": 2.572347266881029e-06,
      "loss": 0.0321,
      "step": 160
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.9365035891532898,
      "learning_rate": 2.6527331189710615e-06,
      "loss": 0.0149,
      "step": 165
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.9742757081985474,
      "learning_rate": 2.7331189710610936e-06,
      "loss": 0.0186,
      "step": 170
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.9915574193000793,
      "learning_rate": 2.813504823151126e-06,
      "loss": 0.0308,
      "step": 175
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.514364719390869,
      "learning_rate": 2.8938906752411576e-06,
      "loss": 0.0288,
      "step": 180
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.717421293258667,
      "learning_rate": 2.97427652733119e-06,
      "loss": 0.0371,
      "step": 185
    },
    {
      "epoch": 0.02,
      "grad_norm": 4.118988990783691,
      "learning_rate": 3.054662379421222e-06,
      "loss": 0.0494,
      "step": 190
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.1433117389678955,
      "learning_rate": 3.135048231511254e-06,
      "loss": 0.0515,
      "step": 195
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.93222975730896,
      "learning_rate": 3.2154340836012865e-06,
      "loss": 0.0276,
      "step": 200
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.6795055270195007,
      "learning_rate": 3.2958199356913185e-06,
      "loss": 0.0318,
      "step": 205
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.950875997543335,
      "learning_rate": 3.376205787781351e-06,
      "loss": 0.0335,
      "step": 210
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.8429243564605713,
      "learning_rate": 3.456591639871383e-06,
      "loss": 0.0434,
      "step": 215
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.2567358016967773,
      "learning_rate": 3.5369774919614154e-06,
      "loss": 0.0324,
      "step": 220
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.407845139503479,
      "learning_rate": 3.617363344051447e-06,
      "loss": 0.0309,
      "step": 225
    },
    {
      "epoch": 0.02,
      "grad_norm": 3.4068539142608643,
      "learning_rate": 3.69774919614148e-06,
      "loss": 0.0454,
      "step": 230
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.8196914196014404,
      "learning_rate": 3.7781350482315114e-06,
      "loss": 0.0255,
      "step": 235
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.0097465515136719,
      "learning_rate": 3.8585209003215434e-06,
      "loss": 0.0263,
      "step": 240
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.995619773864746,
      "learning_rate": 3.938906752411576e-06,
      "loss": 0.0331,
      "step": 245
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.168858528137207,
      "learning_rate": 4.0192926045016075e-06,
      "loss": 0.0416,
      "step": 250
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.8158152103424072,
      "learning_rate": 4.09967845659164e-06,
      "loss": 0.0385,
      "step": 255
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.8158726692199707,
      "learning_rate": 4.180064308681672e-06,
      "loss": 0.0385,
      "step": 260
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.6891674995422363,
      "learning_rate": 4.260450160771704e-06,
      "loss": 0.0182,
      "step": 265
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.5722671747207642,
      "learning_rate": 4.340836012861736e-06,
      "loss": 0.0323,
      "step": 270
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.8213624954223633,
      "learning_rate": 4.421221864951769e-06,
      "loss": 0.0351,
      "step": 275
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.2886526584625244,
      "learning_rate": 4.501607717041801e-06,
      "loss": 0.0401,
      "step": 280
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.6343109607696533,
      "learning_rate": 4.581993569131833e-06,
      "loss": 0.0343,
      "step": 285
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.7094342708587646,
      "learning_rate": 4.662379421221865e-06,
      "loss": 0.0261,
      "step": 290
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.2275326251983643,
      "learning_rate": 4.742765273311897e-06,
      "loss": 0.0183,
      "step": 295
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.9329387545585632,
      "learning_rate": 4.82315112540193e-06,
      "loss": 0.0651,
      "step": 300
    },
    {
      "epoch": 0.02,
      "eval_loss": 0.03527999296784401,
      "eval_pearson_cosine": 0.8237974983947806,
      "eval_pearson_dot": 0.79982187310788,
      "eval_pearson_euclidean": 0.8049934987510711,
      "eval_pearson_manhattan": 0.802979519881207,
      "eval_pearson_max": 0.8237974983947806,
      "eval_runtime": 426.4216,
      "eval_samples_per_second": 1.173,
      "eval_spearman_cosine": 0.8749033396133584,
      "eval_spearman_dot": 0.8475867183468735,
      "eval_spearman_euclidean": 0.8703331453325813,
      "eval_spearman_manhattan": 0.8672266689066755,
      "eval_spearman_max": 0.8749033396133584,
      "eval_steps_per_second": 1.173,
      "step": 300
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.6371708512306213,
      "learning_rate": 4.903536977491961e-06,
      "loss": 0.019,
      "step": 305
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.065721035003662,
      "learning_rate": 4.983922829581994e-06,
      "loss": 0.0193,
      "step": 310
    },
    {
      "epoch": 0.03,
      "grad_norm": 3.1992969512939453,
      "learning_rate": 5.064308681672026e-06,
      "loss": 0.0579,
      "step": 315
    },
    {
      "epoch": 0.03,
      "grad_norm": 2.422349214553833,
      "learning_rate": 5.144694533762058e-06,
      "loss": 0.0342,
      "step": 320
    },
    {
      "epoch": 0.03,
      "grad_norm": 2.09092378616333,
      "learning_rate": 5.22508038585209e-06,
      "loss": 0.0321,
      "step": 325
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.6249920129776,
      "learning_rate": 5.305466237942123e-06,
      "loss": 0.05,
      "step": 330
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.4493677616119385,
      "learning_rate": 5.385852090032154e-06,
      "loss": 0.0371,
      "step": 335
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.1457599401474,
      "learning_rate": 5.466237942122187e-06,
      "loss": 0.0191,
      "step": 340
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.7196341156959534,
      "learning_rate": 5.546623794212219e-06,
      "loss": 0.0653,
      "step": 345
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.7609484791755676,
      "learning_rate": 5.627009646302252e-06,
      "loss": 0.0234,
      "step": 350
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.9668811559677124,
      "learning_rate": 5.707395498392283e-06,
      "loss": 0.043,
      "step": 355
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.4503982067108154,
      "learning_rate": 5.787781350482315e-06,
      "loss": 0.0423,
      "step": 360
    },
    {
      "epoch": 0.03,
      "grad_norm": 2.6571695804595947,
      "learning_rate": 5.868167202572348e-06,
      "loss": 0.0463,
      "step": 365
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.0262566804885864,
      "learning_rate": 5.94855305466238e-06,
      "loss": 0.0362,
      "step": 370
    },
    {
      "epoch": 0.03,
      "grad_norm": 3.1059563159942627,
      "learning_rate": 6.028938906752412e-06,
      "loss": 0.0291,
      "step": 375
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.3710355758666992,
      "learning_rate": 6.109324758842444e-06,
      "loss": 0.0286,
      "step": 380
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.9536277055740356,
      "learning_rate": 6.189710610932477e-06,
      "loss": 0.048,
      "step": 385
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.9812152981758118,
      "learning_rate": 6.270096463022508e-06,
      "loss": 0.0211,
      "step": 390
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.8256497383117676,
      "learning_rate": 6.350482315112541e-06,
      "loss": 0.0248,
      "step": 395
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.5534074306488037,
      "learning_rate": 6.430868167202573e-06,
      "loss": 0.0406,
      "step": 400
    },
    {
      "epoch": 0.03,
      "grad_norm": 2.0551486015319824,
      "learning_rate": 6.511254019292606e-06,
      "loss": 0.0299,
      "step": 405
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.8460248708724976,
      "learning_rate": 6.591639871382637e-06,
      "loss": 0.0182,
      "step": 410
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.7704716920852661,
      "learning_rate": 6.672025723472669e-06,
      "loss": 0.0223,
      "step": 415
    },
    {
      "epoch": 0.03,
      "grad_norm": 8.103775024414062,
      "learning_rate": 6.752411575562702e-06,
      "loss": 0.0296,
      "step": 420
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.514302134513855,
      "learning_rate": 6.832797427652733e-06,
      "loss": 0.0302,
      "step": 425
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.5787880420684814,
      "learning_rate": 6.913183279742766e-06,
      "loss": 0.0229,
      "step": 430
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.478020429611206,
      "learning_rate": 6.993569131832798e-06,
      "loss": 0.03,
      "step": 435
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.8203696608543396,
      "learning_rate": 7.073954983922831e-06,
      "loss": 0.0333,
      "step": 440
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.000648856163025,
      "learning_rate": 7.154340836012862e-06,
      "loss": 0.0177,
      "step": 445
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.5386018753051758,
      "learning_rate": 7.234726688102894e-06,
      "loss": 0.0351,
      "step": 450
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.0292292833328247,
      "learning_rate": 7.315112540192927e-06,
      "loss": 0.0395,
      "step": 455
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.5625340938568115,
      "learning_rate": 7.39549839228296e-06,
      "loss": 0.0299,
      "step": 460
    },
    {
      "epoch": 0.04,
      "grad_norm": 3.6210525035858154,
      "learning_rate": 7.475884244372991e-06,
      "loss": 0.0321,
      "step": 465
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.8112176656723022,
      "learning_rate": 7.556270096463023e-06,
      "loss": 0.0316,
      "step": 470
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.0322524309158325,
      "learning_rate": 7.636655948553056e-06,
      "loss": 0.0253,
      "step": 475
    },
    {
      "epoch": 0.04,
      "grad_norm": 3.308305263519287,
      "learning_rate": 7.717041800643087e-06,
      "loss": 0.0387,
      "step": 480
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.2584173679351807,
      "learning_rate": 7.79742765273312e-06,
      "loss": 0.0353,
      "step": 485
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.6924129128456116,
      "learning_rate": 7.877813504823153e-06,
      "loss": 0.0157,
      "step": 490
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.4199492931365967,
      "learning_rate": 7.958199356913184e-06,
      "loss": 0.0401,
      "step": 495
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.35832083225250244,
      "learning_rate": 8.038585209003215e-06,
      "loss": 0.0227,
      "step": 500
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.6043404340744019,
      "learning_rate": 8.118971061093248e-06,
      "loss": 0.0199,
      "step": 505
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.0943973064422607,
      "learning_rate": 8.19935691318328e-06,
      "loss": 0.0213,
      "step": 510
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.8544149398803711,
      "learning_rate": 8.279742765273314e-06,
      "loss": 0.0226,
      "step": 515
    },
    {
      "epoch": 0.04,
      "grad_norm": 4.021656036376953,
      "learning_rate": 8.360128617363345e-06,
      "loss": 0.0574,
      "step": 520
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.3117215633392334,
      "learning_rate": 8.440514469453378e-06,
      "loss": 0.019,
      "step": 525
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.2881953716278076,
      "learning_rate": 8.520900321543409e-06,
      "loss": 0.0267,
      "step": 530
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.115152597427368,
      "learning_rate": 8.601286173633442e-06,
      "loss": 0.0233,
      "step": 535
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.7420244216918945,
      "learning_rate": 8.681672025723473e-06,
      "loss": 0.0216,
      "step": 540
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.2626619338989258,
      "learning_rate": 8.762057877813506e-06,
      "loss": 0.0182,
      "step": 545
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.5143030881881714,
      "learning_rate": 8.842443729903538e-06,
      "loss": 0.0261,
      "step": 550
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.1002925634384155,
      "learning_rate": 8.92282958199357e-06,
      "loss": 0.022,
      "step": 555
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.9754141569137573,
      "learning_rate": 9.003215434083602e-06,
      "loss": 0.0252,
      "step": 560
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.3871146440505981,
      "learning_rate": 9.083601286173634e-06,
      "loss": 0.0456,
      "step": 565
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.471267580986023,
      "learning_rate": 9.163987138263667e-06,
      "loss": 0.0384,
      "step": 570
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.8994916677474976,
      "learning_rate": 9.244372990353698e-06,
      "loss": 0.036,
      "step": 575
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.7490404844284058,
      "learning_rate": 9.32475884244373e-06,
      "loss": 0.0221,
      "step": 580
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.3971872329711914,
      "learning_rate": 9.405144694533763e-06,
      "loss": 0.0302,
      "step": 585
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.9706817865371704,
      "learning_rate": 9.485530546623795e-06,
      "loss": 0.0291,
      "step": 590
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.0623135566711426,
      "learning_rate": 9.565916398713827e-06,
      "loss": 0.0332,
      "step": 595
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.330348253250122,
      "learning_rate": 9.64630225080386e-06,
      "loss": 0.031,
      "step": 600
    },
    {
      "epoch": 0.05,
      "eval_loss": 0.03203292191028595,
      "eval_pearson_cosine": 0.8607192041230748,
      "eval_pearson_dot": 0.8492954726274867,
      "eval_pearson_euclidean": 0.8425777905216318,
      "eval_pearson_manhattan": 0.8394006469079365,
      "eval_pearson_max": 0.8607192041230748,
      "eval_runtime": 425.6285,
      "eval_samples_per_second": 1.175,
      "eval_spearman_cosine": 0.8971169644678579,
      "eval_spearman_dot": 0.8784386497545991,
      "eval_spearman_euclidean": 0.894069240276961,
      "eval_spearman_manhattan": 0.8926532346129386,
      "eval_spearman_max": 0.8971169644678579,
      "eval_steps_per_second": 1.175,
      "step": 600
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.334221124649048,
      "learning_rate": 9.726688102893891e-06,
      "loss": 0.0273,
      "step": 605
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.7424949407577515,
      "learning_rate": 9.807073954983923e-06,
      "loss": 0.0647,
      "step": 610
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.4894630908966064,
      "learning_rate": 9.887459807073955e-06,
      "loss": 0.0361,
      "step": 615
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.4873441159725189,
      "learning_rate": 9.967845659163988e-06,
      "loss": 0.0264,
      "step": 620
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.5973559617996216,
      "learning_rate": 1.0048231511254021e-05,
      "loss": 0.029,
      "step": 625
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.1919150352478027,
      "learning_rate": 1.0128617363344052e-05,
      "loss": 0.0328,
      "step": 630
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.3969812393188477,
      "learning_rate": 1.0209003215434084e-05,
      "loss": 0.043,
      "step": 635
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.33707332611084,
      "learning_rate": 1.0289389067524116e-05,
      "loss": 0.0448,
      "step": 640
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.3516573905944824,
      "learning_rate": 1.0369774919614148e-05,
      "loss": 0.0328,
      "step": 645
    },
    {
      "epoch": 0.05,
      "grad_norm": 3.014158248901367,
      "learning_rate": 1.045016077170418e-05,
      "loss": 0.0326,
      "step": 650
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.1477378606796265,
      "learning_rate": 1.0530546623794213e-05,
      "loss": 0.025,
      "step": 655
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.8221299648284912,
      "learning_rate": 1.0610932475884246e-05,
      "loss": 0.0425,
      "step": 660
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.68177330493927,
      "learning_rate": 1.0691318327974279e-05,
      "loss": 0.025,
      "step": 665
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.4552159309387207,
      "learning_rate": 1.0771704180064308e-05,
      "loss": 0.0398,
      "step": 670
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.352478265762329,
      "learning_rate": 1.0852090032154341e-05,
      "loss": 0.0155,
      "step": 675
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.9538177251815796,
      "learning_rate": 1.0932475884244374e-05,
      "loss": 0.0299,
      "step": 680
    },
    {
      "epoch": 0.06,
      "grad_norm": 2.0173323154449463,
      "learning_rate": 1.1012861736334405e-05,
      "loss": 0.0268,
      "step": 685
    },
    {
      "epoch": 0.06,
      "grad_norm": 2.0908379554748535,
      "learning_rate": 1.1093247588424438e-05,
      "loss": 0.0308,
      "step": 690
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.6585212349891663,
      "learning_rate": 1.1173633440514471e-05,
      "loss": 0.025,
      "step": 695
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.959600567817688,
      "learning_rate": 1.1254019292604504e-05,
      "loss": 0.0219,
      "step": 700
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.2033023834228516,
      "learning_rate": 1.1334405144694535e-05,
      "loss": 0.0315,
      "step": 705
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.735729455947876,
      "learning_rate": 1.1414790996784566e-05,
      "loss": 0.0249,
      "step": 710
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.850893497467041,
      "learning_rate": 1.14951768488746e-05,
      "loss": 0.0248,
      "step": 715
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.902856171131134,
      "learning_rate": 1.157556270096463e-05,
      "loss": 0.0683,
      "step": 720
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.6975336670875549,
      "learning_rate": 1.1655948553054663e-05,
      "loss": 0.0315,
      "step": 725
    },
    {
      "epoch": 0.06,
      "grad_norm": 3.3982033729553223,
      "learning_rate": 1.1736334405144696e-05,
      "loss": 0.0479,
      "step": 730
    },
    {
      "epoch": 0.06,
      "grad_norm": 2.775419235229492,
      "learning_rate": 1.1816720257234729e-05,
      "loss": 0.068,
      "step": 735
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.9351024627685547,
      "learning_rate": 1.189710610932476e-05,
      "loss": 0.0254,
      "step": 740
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.5352950096130371,
      "learning_rate": 1.1977491961414791e-05,
      "loss": 0.0193,
      "step": 745
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.5528688430786133,
      "learning_rate": 1.2057877813504824e-05,
      "loss": 0.039,
      "step": 750
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.6287634372711182,
      "learning_rate": 1.2138263665594855e-05,
      "loss": 0.017,
      "step": 755
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.0720481872558594,
      "learning_rate": 1.2218649517684888e-05,
      "loss": 0.0261,
      "step": 760
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.8808945417404175,
      "learning_rate": 1.2299035369774921e-05,
      "loss": 0.0224,
      "step": 765
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.175379753112793,
      "learning_rate": 1.2379421221864954e-05,
      "loss": 0.0157,
      "step": 770
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.7339066863059998,
      "learning_rate": 1.2459807073954987e-05,
      "loss": 0.0242,
      "step": 775
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.5107578039169312,
      "learning_rate": 1.2540192926045016e-05,
      "loss": 0.0245,
      "step": 780
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.5132850408554077,
      "learning_rate": 1.2620578778135049e-05,
      "loss": 0.0191,
      "step": 785
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.7692009210586548,
      "learning_rate": 1.2700964630225082e-05,
      "loss": 0.019,
      "step": 790
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.1860913038253784,
      "learning_rate": 1.2781350482315113e-05,
      "loss": 0.0121,
      "step": 795
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.9162280559539795,
      "learning_rate": 1.2861736334405146e-05,
      "loss": 0.0195,
      "step": 800
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.2460323572158813,
      "learning_rate": 1.2942122186495179e-05,
      "loss": 0.0223,
      "step": 805
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.7504421472549438,
      "learning_rate": 1.3022508038585212e-05,
      "loss": 0.0325,
      "step": 810
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.3735928535461426,
      "learning_rate": 1.3102893890675241e-05,
      "loss": 0.0213,
      "step": 815
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.1271485090255737,
      "learning_rate": 1.3183279742765274e-05,
      "loss": 0.0219,
      "step": 820
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.8849589228630066,
      "learning_rate": 1.3263665594855307e-05,
      "loss": 0.0143,
      "step": 825
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.7527520060539246,
      "learning_rate": 1.3344051446945338e-05,
      "loss": 0.0241,
      "step": 830
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.0900169610977173,
      "learning_rate": 1.3424437299035371e-05,
      "loss": 0.0414,
      "step": 835
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.1919902563095093,
      "learning_rate": 1.3504823151125404e-05,
      "loss": 0.0234,
      "step": 840
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.8385984301567078,
      "learning_rate": 1.3585209003215437e-05,
      "loss": 0.0239,
      "step": 845
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.9777655005455017,
      "learning_rate": 1.3665594855305466e-05,
      "loss": 0.0145,
      "step": 850
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.2803690433502197,
      "learning_rate": 1.3745980707395499e-05,
      "loss": 0.0274,
      "step": 855
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.1460819244384766,
      "learning_rate": 1.3826366559485532e-05,
      "loss": 0.0373,
      "step": 860
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.8751824498176575,
      "learning_rate": 1.3906752411575563e-05,
      "loss": 0.0204,
      "step": 865
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.566417932510376,
      "learning_rate": 1.3987138263665596e-05,
      "loss": 0.0139,
      "step": 870
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.6487017869949341,
      "learning_rate": 1.4067524115755629e-05,
      "loss": 0.0458,
      "step": 875
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.9419571161270142,
      "learning_rate": 1.4147909967845662e-05,
      "loss": 0.0215,
      "step": 880
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.7554964423179626,
      "learning_rate": 1.4228295819935693e-05,
      "loss": 0.0219,
      "step": 885
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.240870475769043,
      "learning_rate": 1.4308681672025724e-05,
      "loss": 0.0291,
      "step": 890
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.2490875720977783,
      "learning_rate": 1.4389067524115757e-05,
      "loss": 0.0256,
      "step": 895
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.9301912188529968,
      "learning_rate": 1.4469453376205788e-05,
      "loss": 0.0303,
      "step": 900
    },
    {
      "epoch": 0.07,
      "eval_loss": 0.0409804992377758,
      "eval_pearson_cosine": 0.8657545592954407,
      "eval_pearson_dot": 0.8604614808110218,
      "eval_pearson_euclidean": 0.849931010036339,
      "eval_pearson_manhattan": 0.84985721810182,
      "eval_pearson_max": 0.8657545592954407,
      "eval_runtime": 425.4053,
      "eval_samples_per_second": 1.175,
      "eval_spearman_cosine": 0.9034730538922157,
      "eval_spearman_dot": 0.8936158304633218,
      "eval_spearman_euclidean": 0.9027816591266363,
      "eval_spearman_manhattan": 0.9029234516938067,
      "eval_spearman_max": 0.9034730538922157,
      "eval_steps_per_second": 1.175,
      "step": 900
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.6353436708450317,
      "learning_rate": 1.454983922829582e-05,
      "loss": 0.0306,
      "step": 905
    },
    {
      "epoch": 0.07,
      "grad_norm": 3.5071334838867188,
      "learning_rate": 1.4630225080385854e-05,
      "loss": 0.0529,
      "step": 910
    },
    {
      "epoch": 0.07,
      "grad_norm": 2.511518716812134,
      "learning_rate": 1.4710610932475886e-05,
      "loss": 0.0351,
      "step": 915
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.8150439858436584,
      "learning_rate": 1.479099678456592e-05,
      "loss": 0.0304,
      "step": 920
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.909027636051178,
      "learning_rate": 1.4871382636655949e-05,
      "loss": 0.0306,
      "step": 925
    },
    {
      "epoch": 0.07,
      "grad_norm": 2.117823362350464,
      "learning_rate": 1.4951768488745982e-05,
      "loss": 0.0444,
      "step": 930
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.0820565223693848,
      "learning_rate": 1.5032154340836015e-05,
      "loss": 0.0147,
      "step": 935
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.0721031427383423,
      "learning_rate": 1.5112540192926046e-05,
      "loss": 0.0379,
      "step": 940
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.6273934841156006,
      "learning_rate": 1.5192926045016079e-05,
      "loss": 0.0202,
      "step": 945
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.4175567626953125,
      "learning_rate": 1.527331189710611e-05,
      "loss": 0.0265,
      "step": 950
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.7840615510940552,
      "learning_rate": 1.5353697749196143e-05,
      "loss": 0.0194,
      "step": 955
    },
    {
      "epoch": 0.08,
      "grad_norm": 2.6450772285461426,
      "learning_rate": 1.5434083601286174e-05,
      "loss": 0.0271,
      "step": 960
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.1013175249099731,
      "learning_rate": 1.5514469453376205e-05,
      "loss": 0.035,
      "step": 965
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.6244901418685913,
      "learning_rate": 1.559485530546624e-05,
      "loss": 0.0183,
      "step": 970
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.0193443298339844,
      "learning_rate": 1.567524115755627e-05,
      "loss": 0.0234,
      "step": 975
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.1802617311477661,
      "learning_rate": 1.5755627009646305e-05,
      "loss": 0.0302,
      "step": 980
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.1835167407989502,
      "learning_rate": 1.5836012861736336e-05,
      "loss": 0.0253,
      "step": 985
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.4406228065490723,
      "learning_rate": 1.5916398713826368e-05,
      "loss": 0.0316,
      "step": 990
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.9229764938354492,
      "learning_rate": 1.59967845659164e-05,
      "loss": 0.0408,
      "step": 995
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.6133905649185181,
      "learning_rate": 1.607717041800643e-05,
      "loss": 0.0163,
      "step": 1000
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.388720154762268,
      "learning_rate": 1.6157556270096464e-05,
      "loss": 0.0221,
      "step": 1005
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.9162605404853821,
      "learning_rate": 1.6237942122186496e-05,
      "loss": 0.0292,
      "step": 1010
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.0816845893859863,
      "learning_rate": 1.631832797427653e-05,
      "loss": 0.0284,
      "step": 1015
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.974607527256012,
      "learning_rate": 1.639871382636656e-05,
      "loss": 0.0245,
      "step": 1020
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.7858002185821533,
      "learning_rate": 1.6479099678456592e-05,
      "loss": 0.0208,
      "step": 1025
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.8541093468666077,
      "learning_rate": 1.6559485530546627e-05,
      "loss": 0.0402,
      "step": 1030
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.9254925847053528,
      "learning_rate": 1.6639871382636655e-05,
      "loss": 0.0277,
      "step": 1035
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.1855261325836182,
      "learning_rate": 1.672025723472669e-05,
      "loss": 0.0289,
      "step": 1040
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.1872082948684692,
      "learning_rate": 1.680064308681672e-05,
      "loss": 0.0255,
      "step": 1045
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.9865713119506836,
      "learning_rate": 1.6881028938906755e-05,
      "loss": 0.0331,
      "step": 1050
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.5213806629180908,
      "learning_rate": 1.6961414790996786e-05,
      "loss": 0.0385,
      "step": 1055
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.5804238319396973,
      "learning_rate": 1.7041800643086817e-05,
      "loss": 0.0447,
      "step": 1060
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.2709295749664307,
      "learning_rate": 1.7122186495176852e-05,
      "loss": 0.0184,
      "step": 1065
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.6574845314025879,
      "learning_rate": 1.7202572347266883e-05,
      "loss": 0.0201,
      "step": 1070
    },
    {
      "epoch": 0.09,
      "grad_norm": 2.590877056121826,
      "learning_rate": 1.7282958199356914e-05,
      "loss": 0.0456,
      "step": 1075
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.060797929763794,
      "learning_rate": 1.7363344051446945e-05,
      "loss": 0.0375,
      "step": 1080
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.38614028692245483,
      "learning_rate": 1.744372990353698e-05,
      "loss": 0.0371,
      "step": 1085
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.6158710718154907,
      "learning_rate": 1.752411575562701e-05,
      "loss": 0.0403,
      "step": 1090
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.5008140802383423,
      "learning_rate": 1.7604501607717042e-05,
      "loss": 0.0296,
      "step": 1095
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.41157564520835876,
      "learning_rate": 1.7684887459807077e-05,
      "loss": 0.0255,
      "step": 1100
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.3567136526107788,
      "learning_rate": 1.7765273311897108e-05,
      "loss": 0.0217,
      "step": 1105
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.698455810546875,
      "learning_rate": 1.784565916398714e-05,
      "loss": 0.0188,
      "step": 1110
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.4908955097198486,
      "learning_rate": 1.792604501607717e-05,
      "loss": 0.0188,
      "step": 1115
    },
    {
      "epoch": 0.09,
      "grad_norm": 2.2162156105041504,
      "learning_rate": 1.8006430868167205e-05,
      "loss": 0.02,
      "step": 1120
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.3919101655483246,
      "learning_rate": 1.8086816720257236e-05,
      "loss": 0.0133,
      "step": 1125
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4801468551158905,
      "learning_rate": 1.8167202572347267e-05,
      "loss": 0.0303,
      "step": 1130
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.5365796685218811,
      "learning_rate": 1.8247588424437302e-05,
      "loss": 0.0275,
      "step": 1135
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.777245819568634,
      "learning_rate": 1.8327974276527333e-05,
      "loss": 0.0273,
      "step": 1140
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.5682530403137207,
      "learning_rate": 1.8408360128617364e-05,
      "loss": 0.0261,
      "step": 1145
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.0468765497207642,
      "learning_rate": 1.8488745980707395e-05,
      "loss": 0.0288,
      "step": 1150
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.2527729272842407,
      "learning_rate": 1.856913183279743e-05,
      "loss": 0.0239,
      "step": 1155
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.5002691149711609,
      "learning_rate": 1.864951768488746e-05,
      "loss": 0.0219,
      "step": 1160
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.9675867557525635,
      "learning_rate": 1.8729903536977496e-05,
      "loss": 0.0476,
      "step": 1165
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.8520388007164001,
      "learning_rate": 1.8810289389067527e-05,
      "loss": 0.0258,
      "step": 1170
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.5552634596824646,
      "learning_rate": 1.8890675241157558e-05,
      "loss": 0.0237,
      "step": 1175
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.6992123126983643,
      "learning_rate": 1.897106109324759e-05,
      "loss": 0.0265,
      "step": 1180
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.1183786392211914,
      "learning_rate": 1.905144694533762e-05,
      "loss": 0.027,
      "step": 1185
    },
    {
      "epoch": 0.1,
      "grad_norm": 2.0854079723358154,
      "learning_rate": 1.9131832797427655e-05,
      "loss": 0.0252,
      "step": 1190
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.8095922470092773,
      "learning_rate": 1.9212218649517686e-05,
      "loss": 0.037,
      "step": 1195
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.6434860229492188,
      "learning_rate": 1.929260450160772e-05,
      "loss": 0.0277,
      "step": 1200
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.030386893078684807,
      "eval_pearson_cosine": 0.8622394577970579,
      "eval_pearson_dot": 0.8570136493582737,
      "eval_pearson_euclidean": 0.849571905885398,
      "eval_pearson_manhattan": 0.8505210712126229,
      "eval_pearson_max": 0.8622394577970579,
      "eval_runtime": 425.8405,
      "eval_samples_per_second": 1.174,
      "eval_spearman_cosine": 0.8911063804255216,
      "eval_spearman_dot": 0.8816442945771783,
      "eval_spearman_euclidean": 0.8895569342277367,
      "eval_spearman_manhattan": 0.890007464029856,
      "eval_spearman_max": 0.8911063804255216,
      "eval_steps_per_second": 1.174,
      "step": 1200
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.8423680067062378,
      "learning_rate": 1.9372990353697752e-05,
      "loss": 0.0313,
      "step": 1205
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.2395281791687012,
      "learning_rate": 1.9453376205787783e-05,
      "loss": 0.0297,
      "step": 1210
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.149891972541809,
      "learning_rate": 1.9533762057877814e-05,
      "loss": 0.0208,
      "step": 1215
    },
    {
      "epoch": 0.1,
      "grad_norm": 2.4256904125213623,
      "learning_rate": 1.9614147909967845e-05,
      "loss": 0.0437,
      "step": 1220
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.7631718516349792,
      "learning_rate": 1.969453376205788e-05,
      "loss": 0.0335,
      "step": 1225
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.0927650928497314,
      "learning_rate": 1.977491961414791e-05,
      "loss": 0.0349,
      "step": 1230
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.7310745120048523,
      "learning_rate": 1.9855305466237946e-05,
      "loss": 0.0305,
      "step": 1235
    },
    {
      "epoch": 0.1,
      "grad_norm": 2.624340534210205,
      "learning_rate": 1.9935691318327977e-05,
      "loss": 0.0456,
      "step": 1240
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.09121835231781,
      "learning_rate": 1.999821316894488e-05,
      "loss": 0.0119,
      "step": 1245
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.6216334104537964,
      "learning_rate": 1.998927901366926e-05,
      "loss": 0.0419,
      "step": 1250
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.0775346755981445,
      "learning_rate": 1.998034485839364e-05,
      "loss": 0.0317,
      "step": 1255
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.044417381286621,
      "learning_rate": 1.997141070311802e-05,
      "loss": 0.0245,
      "step": 1260
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.581361174583435,
      "learning_rate": 1.9962476547842403e-05,
      "loss": 0.0246,
      "step": 1265
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.4875144362449646,
      "learning_rate": 1.9953542392566784e-05,
      "loss": 0.0218,
      "step": 1270
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.2498222589492798,
      "learning_rate": 1.9944608237291165e-05,
      "loss": 0.04,
      "step": 1275
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.562926173210144,
      "learning_rate": 1.993567408201555e-05,
      "loss": 0.0268,
      "step": 1280
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.5474754571914673,
      "learning_rate": 1.992673992673993e-05,
      "loss": 0.0514,
      "step": 1285
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.6975357532501221,
      "learning_rate": 1.9917805771464308e-05,
      "loss": 0.0345,
      "step": 1290
    },
    {
      "epoch": 0.1,
      "grad_norm": 2.0237133502960205,
      "learning_rate": 1.9908871616188692e-05,
      "loss": 0.0523,
      "step": 1295
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.604995846748352,
      "learning_rate": 1.9899937460913073e-05,
      "loss": 0.0279,
      "step": 1300
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.3883243799209595,
      "learning_rate": 1.9891003305637454e-05,
      "loss": 0.0227,
      "step": 1305
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.5331462621688843,
      "learning_rate": 1.9882069150361835e-05,
      "loss": 0.0326,
      "step": 1310
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.44855189323425293,
      "learning_rate": 1.9873134995086216e-05,
      "loss": 0.0352,
      "step": 1315
    },
    {
      "epoch": 0.11,
      "grad_norm": 2.213827610015869,
      "learning_rate": 1.9864200839810597e-05,
      "loss": 0.0432,
      "step": 1320
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.5794016718864441,
      "learning_rate": 1.985526668453498e-05,
      "loss": 0.0245,
      "step": 1325
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.6042608618736267,
      "learning_rate": 1.984633252925936e-05,
      "loss": 0.0168,
      "step": 1330
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.935172200202942,
      "learning_rate": 1.983739837398374e-05,
      "loss": 0.0347,
      "step": 1335
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.000755786895752,
      "learning_rate": 1.982846421870812e-05,
      "loss": 0.0302,
      "step": 1340
    },
    {
      "epoch": 0.11,
      "grad_norm": 2.054868698120117,
      "learning_rate": 1.9819530063432503e-05,
      "loss": 0.0622,
      "step": 1345
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.873913049697876,
      "learning_rate": 1.9810595908156884e-05,
      "loss": 0.0178,
      "step": 1350
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.6479474306106567,
      "learning_rate": 1.9801661752881268e-05,
      "loss": 0.0204,
      "step": 1355
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.0146560668945312,
      "learning_rate": 1.979272759760565e-05,
      "loss": 0.0312,
      "step": 1360
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.7326614856719971,
      "learning_rate": 1.978379344233003e-05,
      "loss": 0.0444,
      "step": 1365
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.8839195370674133,
      "learning_rate": 1.977485928705441e-05,
      "loss": 0.0301,
      "step": 1370
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.485784113407135,
      "learning_rate": 1.9765925131778792e-05,
      "loss": 0.0349,
      "step": 1375
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.7656601071357727,
      "learning_rate": 1.9756990976503173e-05,
      "loss": 0.0282,
      "step": 1380
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.6562867164611816,
      "learning_rate": 1.9748056821227554e-05,
      "loss": 0.033,
      "step": 1385
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.971908688545227,
      "learning_rate": 1.9739122665951935e-05,
      "loss": 0.0358,
      "step": 1390
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.4427505731582642,
      "learning_rate": 1.9730188510676316e-05,
      "loss": 0.029,
      "step": 1395
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.177832841873169,
      "learning_rate": 1.9721254355400697e-05,
      "loss": 0.0296,
      "step": 1400
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.5108907222747803,
      "learning_rate": 1.971232020012508e-05,
      "loss": 0.0239,
      "step": 1405
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.7967244386672974,
      "learning_rate": 1.9703386044849463e-05,
      "loss": 0.0336,
      "step": 1410
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.008859395980835,
      "learning_rate": 1.9694451889573844e-05,
      "loss": 0.0197,
      "step": 1415
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.1244670152664185,
      "learning_rate": 1.968551773429822e-05,
      "loss": 0.0407,
      "step": 1420
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.6256637573242188,
      "learning_rate": 1.9676583579022606e-05,
      "loss": 0.0464,
      "step": 1425
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.6180126667022705,
      "learning_rate": 1.9667649423746987e-05,
      "loss": 0.0185,
      "step": 1430
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.8708995580673218,
      "learning_rate": 1.9658715268471368e-05,
      "loss": 0.0206,
      "step": 1435
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.933128833770752,
      "learning_rate": 1.964978111319575e-05,
      "loss": 0.0427,
      "step": 1440
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.588142991065979,
      "learning_rate": 1.964084695792013e-05,
      "loss": 0.0242,
      "step": 1445
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.38248518109321594,
      "learning_rate": 1.963191280264451e-05,
      "loss": 0.0252,
      "step": 1450
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.8211073279380798,
      "learning_rate": 1.9622978647368892e-05,
      "loss": 0.0159,
      "step": 1455
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4323325455188751,
      "learning_rate": 1.9614044492093273e-05,
      "loss": 0.0383,
      "step": 1460
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.7777843475341797,
      "learning_rate": 1.9605110336817658e-05,
      "loss": 0.0399,
      "step": 1465
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.7342023253440857,
      "learning_rate": 1.9596176181542035e-05,
      "loss": 0.0232,
      "step": 1470
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.43286553025245667,
      "learning_rate": 1.9587242026266416e-05,
      "loss": 0.03,
      "step": 1475
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.4914606809616089,
      "learning_rate": 1.95783078709908e-05,
      "loss": 0.0253,
      "step": 1480
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.6109514236450195,
      "learning_rate": 1.9569373715715182e-05,
      "loss": 0.0244,
      "step": 1485
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.099238395690918,
      "learning_rate": 1.9560439560439563e-05,
      "loss": 0.0219,
      "step": 1490
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.672016978263855,
      "learning_rate": 1.9551505405163944e-05,
      "loss": 0.0156,
      "step": 1495
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.9910028576850891,
      "learning_rate": 1.9542571249888325e-05,
      "loss": 0.025,
      "step": 1500
    },
    {
      "epoch": 0.12,
      "eval_loss": 0.029246093705296516,
      "eval_pearson_cosine": 0.8662779121023975,
      "eval_pearson_dot": 0.8609972822698294,
      "eval_pearson_euclidean": 0.8525960773878268,
      "eval_pearson_manhattan": 0.8525524127391249,
      "eval_pearson_max": 0.8662779121023975,
      "eval_runtime": 425.5831,
      "eval_samples_per_second": 1.175,
      "eval_spearman_cosine": 0.8936378145512582,
      "eval_spearman_dot": 0.8876585266341065,
      "eval_spearman_euclidean": 0.891318445273781,
      "eval_spearman_manhattan": 0.889811431245725,
      "eval_spearman_max": 0.8936378145512582,
      "eval_steps_per_second": 1.175,
      "step": 1500
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.7004162073135376,
      "learning_rate": 1.9533637094612706e-05,
      "loss": 0.0248,
      "step": 1505
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.0799355506896973,
      "learning_rate": 1.9524702939337087e-05,
      "loss": 0.0268,
      "step": 1510
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.8577656745910645,
      "learning_rate": 1.9515768784061468e-05,
      "loss": 0.0363,
      "step": 1515
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.811858892440796,
      "learning_rate": 1.950683462878585e-05,
      "loss": 0.0408,
      "step": 1520
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.2648321390151978,
      "learning_rate": 1.949790047351023e-05,
      "loss": 0.0273,
      "step": 1525
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.6332646012306213,
      "learning_rate": 1.948896631823461e-05,
      "loss": 0.0234,
      "step": 1530
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.1494219303131104,
      "learning_rate": 1.9480032162958992e-05,
      "loss": 0.0396,
      "step": 1535
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.5537444353103638,
      "learning_rate": 1.9471098007683377e-05,
      "loss": 0.0392,
      "step": 1540
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.7606208324432373,
      "learning_rate": 1.9462163852407754e-05,
      "loss": 0.0468,
      "step": 1545
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.3266965448856354,
      "learning_rate": 1.945322969713214e-05,
      "loss": 0.0164,
      "step": 1550
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.4910846948623657,
      "learning_rate": 1.944429554185652e-05,
      "loss": 0.0181,
      "step": 1555
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.621263861656189,
      "learning_rate": 1.94353613865809e-05,
      "loss": 0.021,
      "step": 1560
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.8153178691864014,
      "learning_rate": 1.9426427231305282e-05,
      "loss": 0.019,
      "step": 1565
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.9617846012115479,
      "learning_rate": 1.9417493076029663e-05,
      "loss": 0.0186,
      "step": 1570
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.6482746005058289,
      "learning_rate": 1.9408558920754044e-05,
      "loss": 0.0226,
      "step": 1575
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.6985549926757812,
      "learning_rate": 1.9399624765478425e-05,
      "loss": 0.0267,
      "step": 1580
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.9519606828689575,
      "learning_rate": 1.9390690610202806e-05,
      "loss": 0.0187,
      "step": 1585
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.9981301426887512,
      "learning_rate": 1.9381756454927187e-05,
      "loss": 0.0195,
      "step": 1590
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.5453444719314575,
      "learning_rate": 1.937282229965157e-05,
      "loss": 0.0185,
      "step": 1595
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.3481849431991577,
      "learning_rate": 1.936388814437595e-05,
      "loss": 0.0439,
      "step": 1600
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.49602463841438293,
      "learning_rate": 1.935495398910033e-05,
      "loss": 0.0158,
      "step": 1605
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.477524071931839,
      "learning_rate": 1.9346019833824715e-05,
      "loss": 0.0276,
      "step": 1610
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.8374057412147522,
      "learning_rate": 1.9337085678549096e-05,
      "loss": 0.0259,
      "step": 1615
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.9524512887001038,
      "learning_rate": 1.9328151523273477e-05,
      "loss": 0.0403,
      "step": 1620
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.6074572205543518,
      "learning_rate": 1.9319217367997858e-05,
      "loss": 0.0217,
      "step": 1625
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.7999153137207031,
      "learning_rate": 1.931028321272224e-05,
      "loss": 0.0234,
      "step": 1630
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.9614180326461792,
      "learning_rate": 1.930134905744662e-05,
      "loss": 0.0231,
      "step": 1635
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.1887145042419434,
      "learning_rate": 1.9292414902171e-05,
      "loss": 0.0414,
      "step": 1640
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.8795381784439087,
      "learning_rate": 1.9283480746895382e-05,
      "loss": 0.0208,
      "step": 1645
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.3685334920883179,
      "learning_rate": 1.9274546591619766e-05,
      "loss": 0.054,
      "step": 1650
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.7518516778945923,
      "learning_rate": 1.9265612436344144e-05,
      "loss": 0.0274,
      "step": 1655
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.029801607131958,
      "learning_rate": 1.9256678281068525e-05,
      "loss": 0.0174,
      "step": 1660
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.7995644807815552,
      "learning_rate": 1.924774412579291e-05,
      "loss": 0.037,
      "step": 1665
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.5098012685775757,
      "learning_rate": 1.923880997051729e-05,
      "loss": 0.0216,
      "step": 1670
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.5958812832832336,
      "learning_rate": 1.9229875815241668e-05,
      "loss": 0.0142,
      "step": 1675
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.7016432285308838,
      "learning_rate": 1.9220941659966053e-05,
      "loss": 0.0416,
      "step": 1680
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5232900977134705,
      "learning_rate": 1.9212007504690434e-05,
      "loss": 0.0264,
      "step": 1685
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4362630248069763,
      "learning_rate": 1.9203073349414815e-05,
      "loss": 0.022,
      "step": 1690
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.451072335243225,
      "learning_rate": 1.9194139194139196e-05,
      "loss": 0.0262,
      "step": 1695
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5889037251472473,
      "learning_rate": 1.9185205038863577e-05,
      "loss": 0.0267,
      "step": 1700
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.49965912103652954,
      "learning_rate": 1.9176270883587958e-05,
      "loss": 0.0159,
      "step": 1705
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.920385479927063,
      "learning_rate": 1.916733672831234e-05,
      "loss": 0.0357,
      "step": 1710
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5249290466308594,
      "learning_rate": 1.915840257303672e-05,
      "loss": 0.0213,
      "step": 1715
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5141476988792419,
      "learning_rate": 1.91494684177611e-05,
      "loss": 0.0196,
      "step": 1720
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5003548264503479,
      "learning_rate": 1.9140534262485485e-05,
      "loss": 0.0157,
      "step": 1725
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4535447657108307,
      "learning_rate": 1.9131600107209863e-05,
      "loss": 0.019,
      "step": 1730
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4489072263240814,
      "learning_rate": 1.9122665951934247e-05,
      "loss": 0.0168,
      "step": 1735
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5539095997810364,
      "learning_rate": 1.911373179665863e-05,
      "loss": 0.02,
      "step": 1740
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.3125584125518799,
      "learning_rate": 1.910479764138301e-05,
      "loss": 0.0202,
      "step": 1745
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.6378400921821594,
      "learning_rate": 1.909586348610739e-05,
      "loss": 0.0289,
      "step": 1750
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.118848443031311,
      "learning_rate": 1.908692933083177e-05,
      "loss": 0.0242,
      "step": 1755
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.8065999746322632,
      "learning_rate": 1.9077995175556153e-05,
      "loss": 0.0277,
      "step": 1760
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.9087148904800415,
      "learning_rate": 1.9069061020280534e-05,
      "loss": 0.0239,
      "step": 1765
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.6891593933105469,
      "learning_rate": 1.9060126865004915e-05,
      "loss": 0.0217,
      "step": 1770
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.7434559464454651,
      "learning_rate": 1.9051192709729296e-05,
      "loss": 0.0145,
      "step": 1775
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.585472047328949,
      "learning_rate": 1.904225855445368e-05,
      "loss": 0.0287,
      "step": 1780
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.561167597770691,
      "learning_rate": 1.9033324399178058e-05,
      "loss": 0.0206,
      "step": 1785
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.9487882852554321,
      "learning_rate": 1.902439024390244e-05,
      "loss": 0.0212,
      "step": 1790
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.7181324362754822,
      "learning_rate": 1.9015456088626823e-05,
      "loss": 0.0241,
      "step": 1795
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5083106160163879,
      "learning_rate": 1.9006521933351204e-05,
      "loss": 0.0159,
      "step": 1800
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.030221089720726013,
      "eval_pearson_cosine": 0.8857723004603547,
      "eval_pearson_dot": 0.874834502431298,
      "eval_pearson_euclidean": 0.8749650474368084,
      "eval_pearson_manhattan": 0.8746243487782971,
      "eval_pearson_max": 0.8857723004603547,
      "eval_runtime": 426.1522,
      "eval_samples_per_second": 1.173,
      "eval_spearman_cosine": 0.9000108960435841,
      "eval_spearman_dot": 0.8859179436717747,
      "eval_spearman_euclidean": 0.9029959319837279,
      "eval_spearman_manhattan": 0.9019226476905907,
      "eval_spearman_max": 0.9029959319837279,
      "eval_steps_per_second": 1.173,
      "step": 1800
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.43535247445106506,
      "learning_rate": 1.8997587778075582e-05,
      "loss": 0.0153,
      "step": 1805
    },
    {
      "epoch": 0.15,
      "grad_norm": 2.1544151306152344,
      "learning_rate": 1.8988653622799966e-05,
      "loss": 0.0241,
      "step": 1810
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.9339686632156372,
      "learning_rate": 1.8979719467524347e-05,
      "loss": 0.0439,
      "step": 1815
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.8600350618362427,
      "learning_rate": 1.897078531224873e-05,
      "loss": 0.0268,
      "step": 1820
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.7343426942825317,
      "learning_rate": 1.896185115697311e-05,
      "loss": 0.0134,
      "step": 1825
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.237486481666565,
      "learning_rate": 1.895291700169749e-05,
      "loss": 0.0387,
      "step": 1830
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.8042169213294983,
      "learning_rate": 1.8943982846421875e-05,
      "loss": 0.0218,
      "step": 1835
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.5267782211303711,
      "learning_rate": 1.8935048691146253e-05,
      "loss": 0.0277,
      "step": 1840
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.7091891169548035,
      "learning_rate": 1.8926114535870634e-05,
      "loss": 0.0196,
      "step": 1845
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.3862394094467163,
      "learning_rate": 1.8917180380595018e-05,
      "loss": 0.0173,
      "step": 1850
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.8204643726348877,
      "learning_rate": 1.89082462253194e-05,
      "loss": 0.0174,
      "step": 1855
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.6407317519187927,
      "learning_rate": 1.8899312070043777e-05,
      "loss": 0.0152,
      "step": 1860
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.6078154444694519,
      "learning_rate": 1.889037791476816e-05,
      "loss": 0.0287,
      "step": 1865
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.8564119338989258,
      "learning_rate": 1.8881443759492542e-05,
      "loss": 0.0185,
      "step": 1870
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.7499640583992004,
      "learning_rate": 1.8872509604216923e-05,
      "loss": 0.04,
      "step": 1875
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.5337907671928406,
      "learning_rate": 1.8863575448941304e-05,
      "loss": 0.0157,
      "step": 1880
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.525449275970459,
      "learning_rate": 1.8854641293665685e-05,
      "loss": 0.0252,
      "step": 1885
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.069151759147644,
      "learning_rate": 1.8845707138390066e-05,
      "loss": 0.0261,
      "step": 1890
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.5704050660133362,
      "learning_rate": 1.8836772983114447e-05,
      "loss": 0.0228,
      "step": 1895
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.8183658719062805,
      "learning_rate": 1.882783882783883e-05,
      "loss": 0.0293,
      "step": 1900
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.5661863088607788,
      "learning_rate": 1.881890467256321e-05,
      "loss": 0.0214,
      "step": 1905
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.24864475429058075,
      "learning_rate": 1.8809970517287594e-05,
      "loss": 0.0169,
      "step": 1910
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.622249960899353,
      "learning_rate": 1.880103636201197e-05,
      "loss": 0.0267,
      "step": 1915
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.6661076545715332,
      "learning_rate": 1.8792102206736356e-05,
      "loss": 0.0287,
      "step": 1920
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.737390398979187,
      "learning_rate": 1.8783168051460737e-05,
      "loss": 0.0179,
      "step": 1925
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6081177592277527,
      "learning_rate": 1.8774233896185118e-05,
      "loss": 0.0194,
      "step": 1930
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.8305076360702515,
      "learning_rate": 1.87652997409095e-05,
      "loss": 0.0333,
      "step": 1935
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.7024769186973572,
      "learning_rate": 1.875636558563388e-05,
      "loss": 0.0272,
      "step": 1940
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.7568246722221375,
      "learning_rate": 1.874743143035826e-05,
      "loss": 0.0182,
      "step": 1945
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.3228060007095337,
      "learning_rate": 1.8738497275082642e-05,
      "loss": 0.0223,
      "step": 1950
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.272178053855896,
      "learning_rate": 1.8729563119807023e-05,
      "loss": 0.0299,
      "step": 1955
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.30096063017845154,
      "learning_rate": 1.8720628964531404e-05,
      "loss": 0.017,
      "step": 1960
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.5879131555557251,
      "learning_rate": 1.871169480925579e-05,
      "loss": 0.0265,
      "step": 1965
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6623579263687134,
      "learning_rate": 1.8702760653980166e-05,
      "loss": 0.024,
      "step": 1970
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.7402109503746033,
      "learning_rate": 1.8693826498704547e-05,
      "loss": 0.0248,
      "step": 1975
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.46998223662376404,
      "learning_rate": 1.8684892343428932e-05,
      "loss": 0.0367,
      "step": 1980
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.8414738178253174,
      "learning_rate": 1.8675958188153313e-05,
      "loss": 0.0352,
      "step": 1985
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.5672045350074768,
      "learning_rate": 1.866702403287769e-05,
      "loss": 0.0286,
      "step": 1990
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4115860164165497,
      "learning_rate": 1.8658089877602075e-05,
      "loss": 0.0235,
      "step": 1995
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6932913064956665,
      "learning_rate": 1.8649155722326456e-05,
      "loss": 0.0128,
      "step": 2000
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6872785687446594,
      "learning_rate": 1.8640221567050837e-05,
      "loss": 0.0205,
      "step": 2005
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.590305745601654,
      "learning_rate": 1.8631287411775218e-05,
      "loss": 0.0207,
      "step": 2010
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.5067042708396912,
      "learning_rate": 1.86223532564996e-05,
      "loss": 0.038,
      "step": 2015
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.46255382895469666,
      "learning_rate": 1.861341910122398e-05,
      "loss": 0.0237,
      "step": 2020
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.4533158540725708,
      "learning_rate": 1.860448494594836e-05,
      "loss": 0.0197,
      "step": 2025
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.919363796710968,
      "learning_rate": 1.8595550790672742e-05,
      "loss": 0.023,
      "step": 2030
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.42132195830345154,
      "learning_rate": 1.8586616635397127e-05,
      "loss": 0.0119,
      "step": 2035
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.9792714715003967,
      "learning_rate": 1.8577682480121508e-05,
      "loss": 0.022,
      "step": 2040
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.0622044801712036,
      "learning_rate": 1.8568748324845885e-05,
      "loss": 0.0191,
      "step": 2045
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.3776015341281891,
      "learning_rate": 1.855981416957027e-05,
      "loss": 0.0173,
      "step": 2050
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.7663435935974121,
      "learning_rate": 1.855088001429465e-05,
      "loss": 0.0257,
      "step": 2055
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5264787673950195,
      "learning_rate": 1.8541945859019032e-05,
      "loss": 0.023,
      "step": 2060
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.0220805406570435,
      "learning_rate": 1.8533011703743413e-05,
      "loss": 0.0215,
      "step": 2065
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.47214239835739136,
      "learning_rate": 1.8524077548467794e-05,
      "loss": 0.0368,
      "step": 2070
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.40313026309013367,
      "learning_rate": 1.8515143393192175e-05,
      "loss": 0.0199,
      "step": 2075
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.7139260768890381,
      "learning_rate": 1.8506209237916556e-05,
      "loss": 0.0182,
      "step": 2080
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.0911624431610107,
      "learning_rate": 1.8497275082640937e-05,
      "loss": 0.0215,
      "step": 2085
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5224277973175049,
      "learning_rate": 1.8488340927365318e-05,
      "loss": 0.0203,
      "step": 2090
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.6372708678245544,
      "learning_rate": 1.84794067720897e-05,
      "loss": 0.0363,
      "step": 2095
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.0599619150161743,
      "learning_rate": 1.847047261681408e-05,
      "loss": 0.0343,
      "step": 2100
    },
    {
      "epoch": 0.17,
      "eval_loss": 0.023791618645191193,
      "eval_pearson_cosine": 0.8906459396898687,
      "eval_pearson_dot": 0.8866654230263398,
      "eval_pearson_euclidean": 0.8789258913375184,
      "eval_pearson_manhattan": 0.8774504409979331,
      "eval_pearson_max": 0.8906459396898687,
      "eval_runtime": 425.9801,
      "eval_samples_per_second": 1.174,
      "eval_spearman_cosine": 0.9069881079524318,
      "eval_spearman_dot": 0.8997297109188437,
      "eval_spearman_euclidean": 0.9081995367981472,
      "eval_spearman_manhattan": 0.90537923751695,
      "eval_spearman_max": 0.9081995367981472,
      "eval_steps_per_second": 1.174,
      "step": 2100
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.0371475219726562,
      "learning_rate": 1.8461538461538465e-05,
      "loss": 0.0205,
      "step": 2105
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.7635193467140198,
      "learning_rate": 1.8452604306262846e-05,
      "loss": 0.0236,
      "step": 2110
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.4593610167503357,
      "learning_rate": 1.8443670150987227e-05,
      "loss": 0.0223,
      "step": 2115
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.9037809371948242,
      "learning_rate": 1.8434735995711608e-05,
      "loss": 0.0348,
      "step": 2120
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.4961464405059814,
      "learning_rate": 1.842580184043599e-05,
      "loss": 0.0278,
      "step": 2125
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.0911996364593506,
      "learning_rate": 1.841686768516037e-05,
      "loss": 0.0124,
      "step": 2130
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.8265781998634338,
      "learning_rate": 1.840793352988475e-05,
      "loss": 0.0336,
      "step": 2135
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.2838935852050781,
      "learning_rate": 1.8398999374609132e-05,
      "loss": 0.0271,
      "step": 2140
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.8582805395126343,
      "learning_rate": 1.8390065219333513e-05,
      "loss": 0.0466,
      "step": 2145
    },
    {
      "epoch": 0.17,
      "grad_norm": 2.2625269889831543,
      "learning_rate": 1.8381131064057894e-05,
      "loss": 0.0338,
      "step": 2150
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.7238175272941589,
      "learning_rate": 1.8372196908782275e-05,
      "loss": 0.0149,
      "step": 2155
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.4642048478126526,
      "learning_rate": 1.8363262753506656e-05,
      "loss": 0.017,
      "step": 2160
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.6727991700172424,
      "learning_rate": 1.835432859823104e-05,
      "loss": 0.0302,
      "step": 2165
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.1912602186203003,
      "learning_rate": 1.834539444295542e-05,
      "loss": 0.0244,
      "step": 2170
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.8099949955940247,
      "learning_rate": 1.83364602876798e-05,
      "loss": 0.013,
      "step": 2175
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.6302483081817627,
      "learning_rate": 1.8327526132404184e-05,
      "loss": 0.0245,
      "step": 2180
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.44703614711761475,
      "learning_rate": 1.8318591977128565e-05,
      "loss": 0.0179,
      "step": 2185
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4171731472015381,
      "learning_rate": 1.8309657821852946e-05,
      "loss": 0.0155,
      "step": 2190
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.9061782360076904,
      "learning_rate": 1.8300723666577327e-05,
      "loss": 0.0222,
      "step": 2195
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.616051197052002,
      "learning_rate": 1.8291789511301708e-05,
      "loss": 0.0167,
      "step": 2200
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.4685133695602417,
      "learning_rate": 1.828285535602609e-05,
      "loss": 0.0259,
      "step": 2205
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.9447664022445679,
      "learning_rate": 1.827392120075047e-05,
      "loss": 0.0351,
      "step": 2210
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.0720714330673218,
      "learning_rate": 1.826498704547485e-05,
      "loss": 0.0164,
      "step": 2215
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.7534575462341309,
      "learning_rate": 1.8256052890199235e-05,
      "loss": 0.0161,
      "step": 2220
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.8038508892059326,
      "learning_rate": 1.8247118734923613e-05,
      "loss": 0.0205,
      "step": 2225
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.0298678874969482,
      "learning_rate": 1.8238184579647994e-05,
      "loss": 0.019,
      "step": 2230
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.3235474824905396,
      "learning_rate": 1.822925042437238e-05,
      "loss": 0.0265,
      "step": 2235
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.6873222589492798,
      "learning_rate": 1.822031626909676e-05,
      "loss": 0.0258,
      "step": 2240
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.4005903899669647,
      "learning_rate": 1.821138211382114e-05,
      "loss": 0.0185,
      "step": 2245
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.47836175560951233,
      "learning_rate": 1.820244795854552e-05,
      "loss": 0.0229,
      "step": 2250
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.653409481048584,
      "learning_rate": 1.8193513803269903e-05,
      "loss": 0.0155,
      "step": 2255
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.2501707077026367,
      "learning_rate": 1.8184579647994284e-05,
      "loss": 0.0253,
      "step": 2260
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.5146343111991882,
      "learning_rate": 1.8175645492718665e-05,
      "loss": 0.0355,
      "step": 2265
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.7069603204727173,
      "learning_rate": 1.8166711337443046e-05,
      "loss": 0.0245,
      "step": 2270
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.3391192853450775,
      "learning_rate": 1.8157777182167427e-05,
      "loss": 0.0125,
      "step": 2275
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.3187044858932495,
      "learning_rate": 1.8148843026891808e-05,
      "loss": 0.0314,
      "step": 2280
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.3739178478717804,
      "learning_rate": 1.813990887161619e-05,
      "loss": 0.0254,
      "step": 2285
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.9314795732498169,
      "learning_rate": 1.8130974716340573e-05,
      "loss": 0.0264,
      "step": 2290
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.9376474022865295,
      "learning_rate": 1.8122040561064954e-05,
      "loss": 0.0172,
      "step": 2295
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.9479859471321106,
      "learning_rate": 1.8113106405789332e-05,
      "loss": 0.0321,
      "step": 2300
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.8741356730461121,
      "learning_rate": 1.8104172250513716e-05,
      "loss": 0.0216,
      "step": 2305
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.7016437649726868,
      "learning_rate": 1.8095238095238097e-05,
      "loss": 0.0181,
      "step": 2310
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.6568303108215332,
      "learning_rate": 1.808630393996248e-05,
      "loss": 0.0374,
      "step": 2315
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.4468625783920288,
      "learning_rate": 1.807736978468686e-05,
      "loss": 0.021,
      "step": 2320
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.8952460885047913,
      "learning_rate": 1.806843562941124e-05,
      "loss": 0.0215,
      "step": 2325
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.5987122654914856,
      "learning_rate": 1.805950147413562e-05,
      "loss": 0.0182,
      "step": 2330
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.8986082077026367,
      "learning_rate": 1.8050567318860003e-05,
      "loss": 0.0224,
      "step": 2335
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.4435628652572632,
      "learning_rate": 1.8041633163584384e-05,
      "loss": 0.0246,
      "step": 2340
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.5219196081161499,
      "learning_rate": 1.8032699008308765e-05,
      "loss": 0.0222,
      "step": 2345
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.8987548351287842,
      "learning_rate": 1.802376485303315e-05,
      "loss": 0.0416,
      "step": 2350
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.3122328817844391,
      "learning_rate": 1.8014830697757527e-05,
      "loss": 0.0162,
      "step": 2355
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.45380455255508423,
      "learning_rate": 1.8005896542481908e-05,
      "loss": 0.0348,
      "step": 2360
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.7789851427078247,
      "learning_rate": 1.7996962387206292e-05,
      "loss": 0.0207,
      "step": 2365
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.7289481163024902,
      "learning_rate": 1.7988028231930673e-05,
      "loss": 0.0154,
      "step": 2370
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.5287164449691772,
      "learning_rate": 1.7979094076655054e-05,
      "loss": 0.0306,
      "step": 2375
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.236247181892395,
      "learning_rate": 1.7970159921379435e-05,
      "loss": 0.0346,
      "step": 2380
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.6785498857498169,
      "learning_rate": 1.7961225766103816e-05,
      "loss": 0.0191,
      "step": 2385
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.7532624006271362,
      "learning_rate": 1.7952291610828197e-05,
      "loss": 0.0247,
      "step": 2390
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.8474083542823792,
      "learning_rate": 1.794335745555258e-05,
      "loss": 0.028,
      "step": 2395
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.8650068640708923,
      "learning_rate": 1.793442330027696e-05,
      "loss": 0.0234,
      "step": 2400
    },
    {
      "epoch": 0.19,
      "eval_loss": 0.02186727523803711,
      "eval_pearson_cosine": 0.9016899300379176,
      "eval_pearson_dot": 0.8979110419976675,
      "eval_pearson_euclidean": 0.892515905459987,
      "eval_pearson_manhattan": 0.8905747154959989,
      "eval_pearson_max": 0.9016899300379176,
      "eval_runtime": 426.0444,
      "eval_samples_per_second": 1.174,
      "eval_spearman_cosine": 0.9066876267505071,
      "eval_spearman_dot": 0.9012341329365315,
      "eval_spearman_euclidean": 0.9076169104676419,
      "eval_spearman_manhattan": 0.9052689330757322,
      "eval_spearman_max": 0.9076169104676419,
      "eval_steps_per_second": 1.174,
      "step": 2400
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.7486416101455688,
      "learning_rate": 1.7925489145001344e-05,
      "loss": 0.0158,
      "step": 2405
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.20279864966869354,
      "learning_rate": 1.791655498972572e-05,
      "loss": 0.0141,
      "step": 2410
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.4669971466064453,
      "learning_rate": 1.7907620834450103e-05,
      "loss": 0.019,
      "step": 2415
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.6175631880760193,
      "learning_rate": 1.7898686679174487e-05,
      "loss": 0.0148,
      "step": 2420
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.5767547488212585,
      "learning_rate": 1.7889752523898868e-05,
      "loss": 0.0252,
      "step": 2425
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5371144413948059,
      "learning_rate": 1.7880818368623246e-05,
      "loss": 0.0121,
      "step": 2430
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.6145671606063843,
      "learning_rate": 1.787188421334763e-05,
      "loss": 0.0195,
      "step": 2435
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.262970209121704,
      "learning_rate": 1.786295005807201e-05,
      "loss": 0.0298,
      "step": 2440
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.3572503328323364,
      "learning_rate": 1.7854015902796392e-05,
      "loss": 0.0152,
      "step": 2445
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.6049114465713501,
      "learning_rate": 1.7845081747520773e-05,
      "loss": 0.0153,
      "step": 2450
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.40164369344711304,
      "learning_rate": 1.7836147592245154e-05,
      "loss": 0.0171,
      "step": 2455
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.33443671464920044,
      "learning_rate": 1.7827213436969535e-05,
      "loss": 0.0174,
      "step": 2460
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.37140563130378723,
      "learning_rate": 1.7818279281693916e-05,
      "loss": 0.0147,
      "step": 2465
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.44236427545547485,
      "learning_rate": 1.7809345126418297e-05,
      "loss": 0.019,
      "step": 2470
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.1675889492034912,
      "learning_rate": 1.7800410971142682e-05,
      "loss": 0.0196,
      "step": 2475
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.7172742486000061,
      "learning_rate": 1.7791476815867063e-05,
      "loss": 0.0262,
      "step": 2480
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.9804342985153198,
      "learning_rate": 1.778254266059144e-05,
      "loss": 0.0337,
      "step": 2485
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.989030361175537,
      "learning_rate": 1.7773608505315825e-05,
      "loss": 0.0238,
      "step": 2490
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.754467785358429,
      "learning_rate": 1.7764674350040206e-05,
      "loss": 0.0224,
      "step": 2495
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.9083816409111023,
      "learning_rate": 1.7755740194764587e-05,
      "loss": 0.0165,
      "step": 2500
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.5486024022102356,
      "learning_rate": 1.7746806039488968e-05,
      "loss": 0.0173,
      "step": 2505
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.4576135575771332,
      "learning_rate": 1.773787188421335e-05,
      "loss": 0.0177,
      "step": 2510
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.9265009760856628,
      "learning_rate": 1.772893772893773e-05,
      "loss": 0.0216,
      "step": 2515
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.506006121635437,
      "learning_rate": 1.772000357366211e-05,
      "loss": 0.0135,
      "step": 2520
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.2627745270729065,
      "learning_rate": 1.7711069418386492e-05,
      "loss": 0.0244,
      "step": 2525
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.7158712148666382,
      "learning_rate": 1.7702135263110873e-05,
      "loss": 0.0238,
      "step": 2530
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.8237607479095459,
      "learning_rate": 1.7693201107835258e-05,
      "loss": 0.0272,
      "step": 2535
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.266671359539032,
      "learning_rate": 1.7684266952559635e-05,
      "loss": 0.0202,
      "step": 2540
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.6004403829574585,
      "learning_rate": 1.7675332797284016e-05,
      "loss": 0.0414,
      "step": 2545
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.9549585580825806,
      "learning_rate": 1.76663986420084e-05,
      "loss": 0.0209,
      "step": 2550
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.9587855935096741,
      "learning_rate": 1.7657464486732782e-05,
      "loss": 0.0239,
      "step": 2555
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5804222226142883,
      "learning_rate": 1.7648530331457163e-05,
      "loss": 0.0183,
      "step": 2560
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.506087601184845,
      "learning_rate": 1.7639596176181544e-05,
      "loss": 0.0202,
      "step": 2565
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.8496732711791992,
      "learning_rate": 1.7630662020905925e-05,
      "loss": 0.0181,
      "step": 2570
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.3317871391773224,
      "learning_rate": 1.7621727865630306e-05,
      "loss": 0.0218,
      "step": 2575
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.3938535451889038,
      "learning_rate": 1.7612793710354687e-05,
      "loss": 0.0112,
      "step": 2580
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.3293476104736328,
      "learning_rate": 1.7603859555079068e-05,
      "loss": 0.0181,
      "step": 2585
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.28195351362228394,
      "learning_rate": 1.7594925399803452e-05,
      "loss": 0.0135,
      "step": 2590
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.7966617941856384,
      "learning_rate": 1.758599124452783e-05,
      "loss": 0.0145,
      "step": 2595
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5484746694564819,
      "learning_rate": 1.757705708925221e-05,
      "loss": 0.0207,
      "step": 2600
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.576448380947113,
      "learning_rate": 1.7568122933976596e-05,
      "loss": 0.0127,
      "step": 2605
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.28207531571388245,
      "learning_rate": 1.7559188778700977e-05,
      "loss": 0.0132,
      "step": 2610
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.29697030782699585,
      "learning_rate": 1.7550254623425354e-05,
      "loss": 0.0127,
      "step": 2615
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.122375249862671,
      "learning_rate": 1.754132046814974e-05,
      "loss": 0.0297,
      "step": 2620
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5798119902610779,
      "learning_rate": 1.753238631287412e-05,
      "loss": 0.0171,
      "step": 2625
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.6546443104743958,
      "learning_rate": 1.75234521575985e-05,
      "loss": 0.0436,
      "step": 2630
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.8825979828834534,
      "learning_rate": 1.7514518002322882e-05,
      "loss": 0.0138,
      "step": 2635
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.46849188208580017,
      "learning_rate": 1.7505583847047263e-05,
      "loss": 0.0328,
      "step": 2640
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.324487566947937,
      "learning_rate": 1.7496649691771644e-05,
      "loss": 0.0241,
      "step": 2645
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.5214132070541382,
      "learning_rate": 1.7487715536496025e-05,
      "loss": 0.0344,
      "step": 2650
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.791459858417511,
      "learning_rate": 1.7478781381220406e-05,
      "loss": 0.0167,
      "step": 2655
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.888538658618927,
      "learning_rate": 1.746984722594479e-05,
      "loss": 0.0216,
      "step": 2660
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.31034013628959656,
      "learning_rate": 1.746091307066917e-05,
      "loss": 0.0327,
      "step": 2665
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.4363711476325989,
      "learning_rate": 1.745197891539355e-05,
      "loss": 0.0203,
      "step": 2670
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.0624545812606812,
      "learning_rate": 1.7443044760117934e-05,
      "loss": 0.0216,
      "step": 2675
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.4411804676055908,
      "learning_rate": 1.7434110604842315e-05,
      "loss": 0.0294,
      "step": 2680
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.7424313426017761,
      "learning_rate": 1.7425176449566696e-05,
      "loss": 0.0233,
      "step": 2685
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.6678118705749512,
      "learning_rate": 1.7416242294291077e-05,
      "loss": 0.0129,
      "step": 2690
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.43497857451438904,
      "learning_rate": 1.7407308139015458e-05,
      "loss": 0.0272,
      "step": 2695
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.3903897702693939,
      "learning_rate": 1.739837398373984e-05,
      "loss": 0.0209,
      "step": 2700
    },
    {
      "epoch": 0.22,
      "eval_loss": 0.027375079691410065,
      "eval_pearson_cosine": 0.9001051843718942,
      "eval_pearson_dot": 0.9006651139179915,
      "eval_pearson_euclidean": 0.8930651682478097,
      "eval_pearson_manhattan": 0.891765654050478,
      "eval_pearson_max": 0.9006651139179915,
      "eval_runtime": 426.2183,
      "eval_samples_per_second": 1.173,
      "eval_spearman_cosine": 0.9177515750063,
      "eval_spearman_dot": 0.9167129468517873,
      "eval_spearman_euclidean": 0.9167275389101556,
      "eval_spearman_manhattan": 0.9159309277237108,
      "eval_spearman_max": 0.9177515750063,
      "eval_steps_per_second": 1.173,
      "step": 2700
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.3496129512786865,
      "learning_rate": 1.738943982846422e-05,
      "loss": 0.0259,
      "step": 2705
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.1606944799423218,
      "learning_rate": 1.73805056731886e-05,
      "loss": 0.0181,
      "step": 2710
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5323413610458374,
      "learning_rate": 1.7371571517912982e-05,
      "loss": 0.0137,
      "step": 2715
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5756454467773438,
      "learning_rate": 1.7362637362637363e-05,
      "loss": 0.0181,
      "step": 2720
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.3727392256259918,
      "learning_rate": 1.7353703207361744e-05,
      "loss": 0.0123,
      "step": 2725
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.441544771194458,
      "learning_rate": 1.7344769052086125e-05,
      "loss": 0.0157,
      "step": 2730
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.6806672811508179,
      "learning_rate": 1.733583489681051e-05,
      "loss": 0.0196,
      "step": 2735
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5537866353988647,
      "learning_rate": 1.732690074153489e-05,
      "loss": 0.0163,
      "step": 2740
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5595599412918091,
      "learning_rate": 1.731796658625927e-05,
      "loss": 0.0229,
      "step": 2745
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.48301759362220764,
      "learning_rate": 1.7309032430983652e-05,
      "loss": 0.0197,
      "step": 2750
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.2349361181259155,
      "learning_rate": 1.7300098275708034e-05,
      "loss": 0.0198,
      "step": 2755
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.3702278137207031,
      "learning_rate": 1.7291164120432415e-05,
      "loss": 0.0165,
      "step": 2760
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.5885152220726013,
      "learning_rate": 1.7282229965156796e-05,
      "loss": 0.0126,
      "step": 2765
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.2845672369003296,
      "learning_rate": 1.7273295809881177e-05,
      "loss": 0.0232,
      "step": 2770
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.0059493780136108,
      "learning_rate": 1.7264361654605558e-05,
      "loss": 0.0102,
      "step": 2775
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.621356189250946,
      "learning_rate": 1.725542749932994e-05,
      "loss": 0.0149,
      "step": 2780
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.8701572418212891,
      "learning_rate": 1.724649334405432e-05,
      "loss": 0.0147,
      "step": 2785
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.35073745250701904,
      "learning_rate": 1.7237559188778704e-05,
      "loss": 0.0176,
      "step": 2790
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.6895755529403687,
      "learning_rate": 1.7228625033503085e-05,
      "loss": 0.0212,
      "step": 2795
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.9424052834510803,
      "learning_rate": 1.7219690878227463e-05,
      "loss": 0.0213,
      "step": 2800
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.45782536268234253,
      "learning_rate": 1.7210756722951847e-05,
      "loss": 0.0122,
      "step": 2805
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.9347144365310669,
      "learning_rate": 1.720182256767623e-05,
      "loss": 0.0221,
      "step": 2810
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4957163333892822,
      "learning_rate": 1.719288841240061e-05,
      "loss": 0.0154,
      "step": 2815
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.40040886402130127,
      "learning_rate": 1.718395425712499e-05,
      "loss": 0.018,
      "step": 2820
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4707423150539398,
      "learning_rate": 1.717502010184937e-05,
      "loss": 0.0122,
      "step": 2825
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.6660599708557129,
      "learning_rate": 1.7166085946573752e-05,
      "loss": 0.0103,
      "step": 2830
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5606631636619568,
      "learning_rate": 1.7157151791298134e-05,
      "loss": 0.0246,
      "step": 2835
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.0515074729919434,
      "learning_rate": 1.7148217636022515e-05,
      "loss": 0.0173,
      "step": 2840
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.8896968364715576,
      "learning_rate": 1.71392834807469e-05,
      "loss": 0.0357,
      "step": 2845
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.8183419108390808,
      "learning_rate": 1.7130349325471277e-05,
      "loss": 0.0125,
      "step": 2850
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.28626278042793274,
      "learning_rate": 1.7121415170195658e-05,
      "loss": 0.011,
      "step": 2855
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.5537832975387573,
      "learning_rate": 1.7112481014920042e-05,
      "loss": 0.0155,
      "step": 2860
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.3938451111316681,
      "learning_rate": 1.7103546859644423e-05,
      "loss": 0.0163,
      "step": 2865
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.27609241008758545,
      "learning_rate": 1.7094612704368804e-05,
      "loss": 0.0193,
      "step": 2870
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.3002891540527344,
      "learning_rate": 1.7085678549093185e-05,
      "loss": 0.0142,
      "step": 2875
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.8881809711456299,
      "learning_rate": 1.7076744393817566e-05,
      "loss": 0.0296,
      "step": 2880
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.99460768699646,
      "learning_rate": 1.7067810238541947e-05,
      "loss": 0.0157,
      "step": 2885
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.14095938205719,
      "learning_rate": 1.705887608326633e-05,
      "loss": 0.0164,
      "step": 2890
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.2840583026409149,
      "learning_rate": 1.704994192799071e-05,
      "loss": 0.0139,
      "step": 2895
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.3837971091270447,
      "learning_rate": 1.704100777271509e-05,
      "loss": 0.016,
      "step": 2900
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.4202588200569153,
      "learning_rate": 1.703207361743947e-05,
      "loss": 0.0161,
      "step": 2905
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.9836485385894775,
      "learning_rate": 1.7023139462163853e-05,
      "loss": 0.0336,
      "step": 2910
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.28357017040252686,
      "learning_rate": 1.7014205306888234e-05,
      "loss": 0.0151,
      "step": 2915
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.3464276194572449,
      "learning_rate": 1.7005271151612618e-05,
      "loss": 0.0151,
      "step": 2920
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.6251973509788513,
      "learning_rate": 1.6996336996336996e-05,
      "loss": 0.0275,
      "step": 2925
    },
    {
      "epoch": 0.24,
      "grad_norm": 2.1420838832855225,
      "learning_rate": 1.698740284106138e-05,
      "loss": 0.0339,
      "step": 2930
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.43490684032440186,
      "learning_rate": 1.697846868578576e-05,
      "loss": 0.0124,
      "step": 2935
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.576763391494751,
      "learning_rate": 1.6969534530510142e-05,
      "loss": 0.0162,
      "step": 2940
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.8549007773399353,
      "learning_rate": 1.6960600375234523e-05,
      "loss": 0.0217,
      "step": 2945
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.3743274509906769,
      "learning_rate": 1.6951666219958904e-05,
      "loss": 0.0165,
      "step": 2950
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.7152208089828491,
      "learning_rate": 1.6942732064683285e-05,
      "loss": 0.0158,
      "step": 2955
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5613319873809814,
      "learning_rate": 1.6933797909407666e-05,
      "loss": 0.0232,
      "step": 2960
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.3946327269077301,
      "learning_rate": 1.6924863754132047e-05,
      "loss": 0.0156,
      "step": 2965
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5846483707427979,
      "learning_rate": 1.691592959885643e-05,
      "loss": 0.0118,
      "step": 2970
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.6644983291625977,
      "learning_rate": 1.6906995443580813e-05,
      "loss": 0.0156,
      "step": 2975
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.7113365530967712,
      "learning_rate": 1.689806128830519e-05,
      "loss": 0.0131,
      "step": 2980
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.7063427567481995,
      "learning_rate": 1.688912713302957e-05,
      "loss": 0.0196,
      "step": 2985
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.8692206740379333,
      "learning_rate": 1.6880192977753956e-05,
      "loss": 0.02,
      "step": 2990
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.6556200385093689,
      "learning_rate": 1.6871258822478337e-05,
      "loss": 0.0273,
      "step": 2995
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.9636979103088379,
      "learning_rate": 1.6862324667202718e-05,
      "loss": 0.0192,
      "step": 3000
    },
    {
      "epoch": 0.24,
      "eval_loss": 0.0198129303753376,
      "eval_pearson_cosine": 0.9064648118738065,
      "eval_pearson_dot": 0.9033815480817863,
      "eval_pearson_euclidean": 0.8939116848670458,
      "eval_pearson_manhattan": 0.8927147647429707,
      "eval_pearson_max": 0.9064648118738065,
      "eval_runtime": 426.1029,
      "eval_samples_per_second": 1.173,
      "eval_spearman_cosine": 0.9222772251089003,
      "eval_spearman_dot": 0.9168591554366217,
      "eval_spearman_euclidean": 0.9222438169752679,
      "eval_spearman_manhattan": 0.9220269521078085,
      "eval_spearman_max": 0.9222772251089003,
      "eval_steps_per_second": 1.173,
      "step": 3000
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5286144614219666,
      "learning_rate": 1.68533905119271e-05,
      "loss": 0.0228,
      "step": 3005
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.7542079091072083,
      "learning_rate": 1.684445635665148e-05,
      "loss": 0.0156,
      "step": 3010
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.6064092516899109,
      "learning_rate": 1.683552220137586e-05,
      "loss": 0.0258,
      "step": 3015
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.7690675258636475,
      "learning_rate": 1.6826588046100242e-05,
      "loss": 0.0201,
      "step": 3020
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4456731975078583,
      "learning_rate": 1.6817653890824623e-05,
      "loss": 0.0159,
      "step": 3025
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.6881704330444336,
      "learning_rate": 1.6808719735549008e-05,
      "loss": 0.0191,
      "step": 3030
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.515715479850769,
      "learning_rate": 1.6799785580273385e-05,
      "loss": 0.0272,
      "step": 3035
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.9520334601402283,
      "learning_rate": 1.6790851424997766e-05,
      "loss": 0.0164,
      "step": 3040
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.4020353555679321,
      "learning_rate": 1.678191726972215e-05,
      "loss": 0.0224,
      "step": 3045
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.868922472000122,
      "learning_rate": 1.6772983114446532e-05,
      "loss": 0.0343,
      "step": 3050
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5014618635177612,
      "learning_rate": 1.676404895917091e-05,
      "loss": 0.0318,
      "step": 3055
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.820292592048645,
      "learning_rate": 1.6755114803895294e-05,
      "loss": 0.0264,
      "step": 3060
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.6107310056686401,
      "learning_rate": 1.6746180648619675e-05,
      "loss": 0.0188,
      "step": 3065
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.3835718631744385,
      "learning_rate": 1.6737246493344056e-05,
      "loss": 0.012,
      "step": 3070
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.46898066997528076,
      "learning_rate": 1.6728312338068437e-05,
      "loss": 0.0333,
      "step": 3075
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.6556499004364014,
      "learning_rate": 1.6719378182792818e-05,
      "loss": 0.0184,
      "step": 3080
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.8477081060409546,
      "learning_rate": 1.67104440275172e-05,
      "loss": 0.0327,
      "step": 3085
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.6761387586593628,
      "learning_rate": 1.670150987224158e-05,
      "loss": 0.0178,
      "step": 3090
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.4892427623271942,
      "learning_rate": 1.669257571696596e-05,
      "loss": 0.0266,
      "step": 3095
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.7155781388282776,
      "learning_rate": 1.6683641561690342e-05,
      "loss": 0.0242,
      "step": 3100
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.25152644515037537,
      "learning_rate": 1.6674707406414727e-05,
      "loss": 0.0102,
      "step": 3105
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.8297644853591919,
      "learning_rate": 1.6665773251139104e-05,
      "loss": 0.0209,
      "step": 3110
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.36103129386901855,
      "learning_rate": 1.665683909586349e-05,
      "loss": 0.0166,
      "step": 3115
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.8184224367141724,
      "learning_rate": 1.664790494058787e-05,
      "loss": 0.0215,
      "step": 3120
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5235499739646912,
      "learning_rate": 1.663897078531225e-05,
      "loss": 0.0223,
      "step": 3125
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.6001405715942383,
      "learning_rate": 1.6630036630036632e-05,
      "loss": 0.0177,
      "step": 3130
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5021240711212158,
      "learning_rate": 1.6621102474761013e-05,
      "loss": 0.0199,
      "step": 3135
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.2629213035106659,
      "learning_rate": 1.6612168319485394e-05,
      "loss": 0.0375,
      "step": 3140
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.6333101987838745,
      "learning_rate": 1.6603234164209775e-05,
      "loss": 0.0138,
      "step": 3145
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5239858627319336,
      "learning_rate": 1.6594300008934156e-05,
      "loss": 0.0211,
      "step": 3150
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5263399481773376,
      "learning_rate": 1.6585365853658537e-05,
      "loss": 0.0281,
      "step": 3155
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.7547926902770996,
      "learning_rate": 1.657643169838292e-05,
      "loss": 0.0325,
      "step": 3160
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.098939061164856,
      "learning_rate": 1.65674975431073e-05,
      "loss": 0.0136,
      "step": 3165
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5231286883354187,
      "learning_rate": 1.655856338783168e-05,
      "loss": 0.0241,
      "step": 3170
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.6786376237869263,
      "learning_rate": 1.6549629232556064e-05,
      "loss": 0.0264,
      "step": 3175
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.402637243270874,
      "learning_rate": 1.6540695077280446e-05,
      "loss": 0.0214,
      "step": 3180
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.169063687324524,
      "learning_rate": 1.6531760922004823e-05,
      "loss": 0.038,
      "step": 3185
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.9597774147987366,
      "learning_rate": 1.6522826766729208e-05,
      "loss": 0.0289,
      "step": 3190
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5611427426338196,
      "learning_rate": 1.651389261145359e-05,
      "loss": 0.0155,
      "step": 3195
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.9891091585159302,
      "learning_rate": 1.650495845617797e-05,
      "loss": 0.0261,
      "step": 3200
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.445168137550354,
      "learning_rate": 1.649602430090235e-05,
      "loss": 0.0192,
      "step": 3205
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.0663169622421265,
      "learning_rate": 1.6487090145626732e-05,
      "loss": 0.0252,
      "step": 3210
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.9698647856712341,
      "learning_rate": 1.6478155990351116e-05,
      "loss": 0.0223,
      "step": 3215
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.3728877305984497,
      "learning_rate": 1.6469221835075494e-05,
      "loss": 0.0194,
      "step": 3220
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.654509961605072,
      "learning_rate": 1.6460287679799875e-05,
      "loss": 0.0255,
      "step": 3225
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.8245360851287842,
      "learning_rate": 1.645135352452426e-05,
      "loss": 0.0151,
      "step": 3230
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.6417776942253113,
      "learning_rate": 1.644241936924864e-05,
      "loss": 0.0153,
      "step": 3235
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.23088638484477997,
      "learning_rate": 1.6433485213973018e-05,
      "loss": 0.0258,
      "step": 3240
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.4798796474933624,
      "learning_rate": 1.6424551058697402e-05,
      "loss": 0.0152,
      "step": 3245
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.42295366525650024,
      "learning_rate": 1.6415616903421783e-05,
      "loss": 0.0153,
      "step": 3250
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.3053552210330963,
      "learning_rate": 1.6406682748146165e-05,
      "loss": 0.0098,
      "step": 3255
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.37020963430404663,
      "learning_rate": 1.6397748592870546e-05,
      "loss": 0.0151,
      "step": 3260
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5717061758041382,
      "learning_rate": 1.6388814437594927e-05,
      "loss": 0.0157,
      "step": 3265
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5504093766212463,
      "learning_rate": 1.6379880282319308e-05,
      "loss": 0.0397,
      "step": 3270
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.6604989767074585,
      "learning_rate": 1.637094612704369e-05,
      "loss": 0.0158,
      "step": 3275
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.7319514155387878,
      "learning_rate": 1.636201197176807e-05,
      "loss": 0.0118,
      "step": 3280
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5383831262588501,
      "learning_rate": 1.6353077816492454e-05,
      "loss": 0.0196,
      "step": 3285
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.9546700716018677,
      "learning_rate": 1.6344143661216835e-05,
      "loss": 0.0204,
      "step": 3290
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.929811954498291,
      "learning_rate": 1.6335209505941213e-05,
      "loss": 0.0286,
      "step": 3295
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.545769453048706,
      "learning_rate": 1.6326275350665597e-05,
      "loss": 0.0178,
      "step": 3300
    },
    {
      "epoch": 0.27,
      "eval_loss": 0.02022329904139042,
      "eval_pearson_cosine": 0.9200674333472122,
      "eval_pearson_dot": 0.9176731158869308,
      "eval_pearson_euclidean": 0.9061361146439217,
      "eval_pearson_manhattan": 0.9047424977345989,
      "eval_pearson_max": 0.9200674333472122,
      "eval_runtime": 426.1218,
      "eval_samples_per_second": 1.173,
      "eval_spearman_cosine": 0.9299045516182064,
      "eval_spearman_dot": 0.9254067416269663,
      "eval_spearman_euclidean": 0.9302610970443881,
      "eval_spearman_manhattan": 0.9284666578666315,
      "eval_spearman_max": 0.9302610970443881,
      "eval_steps_per_second": 1.173,
      "step": 3300
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5426648259162903,
      "learning_rate": 1.6317341195389978e-05,
      "loss": 0.0123,
      "step": 3305
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.4093814492225647,
      "learning_rate": 1.630840704011436e-05,
      "loss": 0.0176,
      "step": 3310
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.9546383619308472,
      "learning_rate": 1.629947288483874e-05,
      "loss": 0.0195,
      "step": 3315
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.6893852949142456,
      "learning_rate": 1.629053872956312e-05,
      "loss": 0.0253,
      "step": 3320
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.3756393790245056,
      "learning_rate": 1.6281604574287502e-05,
      "loss": 0.0106,
      "step": 3325
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.563850998878479,
      "learning_rate": 1.6272670419011883e-05,
      "loss": 0.017,
      "step": 3330
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.3767915964126587,
      "learning_rate": 1.6263736263736265e-05,
      "loss": 0.0231,
      "step": 3335
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.3786735236644745,
      "learning_rate": 1.6254802108460646e-05,
      "loss": 0.0121,
      "step": 3340
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.2710884213447571,
      "learning_rate": 1.624586795318503e-05,
      "loss": 0.0121,
      "step": 3345
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.4059496819972992,
      "learning_rate": 1.6236933797909408e-05,
      "loss": 0.0138,
      "step": 3350
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5236977338790894,
      "learning_rate": 1.622799964263379e-05,
      "loss": 0.023,
      "step": 3355
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.2646647691726685,
      "learning_rate": 1.6219065487358173e-05,
      "loss": 0.026,
      "step": 3360
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.5769827365875244,
      "learning_rate": 1.6210131332082554e-05,
      "loss": 0.0145,
      "step": 3365
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.8552572727203369,
      "learning_rate": 1.6201197176806932e-05,
      "loss": 0.0132,
      "step": 3370
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.690263271331787,
      "learning_rate": 1.6192263021531316e-05,
      "loss": 0.04,
      "step": 3375
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.6928017735481262,
      "learning_rate": 1.6183328866255697e-05,
      "loss": 0.0184,
      "step": 3380
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.31537023186683655,
      "learning_rate": 1.6174394710980078e-05,
      "loss": 0.0156,
      "step": 3385
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.826603889465332,
      "learning_rate": 1.616546055570446e-05,
      "loss": 0.0253,
      "step": 3390
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.0481098890304565,
      "learning_rate": 1.615652640042884e-05,
      "loss": 0.019,
      "step": 3395
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.719532310962677,
      "learning_rate": 1.614759224515322e-05,
      "loss": 0.0152,
      "step": 3400
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.7920011281967163,
      "learning_rate": 1.6138658089877602e-05,
      "loss": 0.0142,
      "step": 3405
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.9310560822486877,
      "learning_rate": 1.6129723934601983e-05,
      "loss": 0.0187,
      "step": 3410
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.4473055303096771,
      "learning_rate": 1.6120789779326368e-05,
      "loss": 0.0317,
      "step": 3415
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.3835084140300751,
      "learning_rate": 1.611185562405075e-05,
      "loss": 0.0143,
      "step": 3420
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.6586607098579407,
      "learning_rate": 1.6102921468775127e-05,
      "loss": 0.027,
      "step": 3425
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.510007917881012,
      "learning_rate": 1.609398731349951e-05,
      "loss": 0.0213,
      "step": 3430
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.7229037284851074,
      "learning_rate": 1.6085053158223892e-05,
      "loss": 0.014,
      "step": 3435
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.35988929867744446,
      "learning_rate": 1.6076119002948273e-05,
      "loss": 0.0166,
      "step": 3440
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.9715976119041443,
      "learning_rate": 1.6067184847672654e-05,
      "loss": 0.0202,
      "step": 3445
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.5804651975631714,
      "learning_rate": 1.6058250692397035e-05,
      "loss": 0.0177,
      "step": 3450
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.6329138278961182,
      "learning_rate": 1.6049316537121416e-05,
      "loss": 0.0184,
      "step": 3455
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.46104130148887634,
      "learning_rate": 1.6040382381845797e-05,
      "loss": 0.0208,
      "step": 3460
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.040407657623291,
      "learning_rate": 1.6031448226570178e-05,
      "loss": 0.0281,
      "step": 3465
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.7342686057090759,
      "learning_rate": 1.6022514071294563e-05,
      "loss": 0.0163,
      "step": 3470
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.7275007367134094,
      "learning_rate": 1.601357991601894e-05,
      "loss": 0.0193,
      "step": 3475
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.30385661125183105,
      "learning_rate": 1.600464576074332e-05,
      "loss": 0.0153,
      "step": 3480
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.5035312175750732,
      "learning_rate": 1.5995711605467706e-05,
      "loss": 0.0189,
      "step": 3485
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.2530502676963806,
      "learning_rate": 1.5986777450192087e-05,
      "loss": 0.017,
      "step": 3490
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.068692922592163,
      "learning_rate": 1.5977843294916468e-05,
      "loss": 0.0178,
      "step": 3495
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.40073007345199585,
      "learning_rate": 1.596890913964085e-05,
      "loss": 0.0165,
      "step": 3500
    },
    {
      "epoch": 0.28,
      "grad_norm": 6.783413887023926,
      "learning_rate": 1.595997498436523e-05,
      "loss": 0.0235,
      "step": 3505
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.46688321232795715,
      "learning_rate": 1.595104082908961e-05,
      "loss": 0.0188,
      "step": 3510
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.6832770705223083,
      "learning_rate": 1.5942106673813992e-05,
      "loss": 0.0202,
      "step": 3515
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.39423632621765137,
      "learning_rate": 1.5933172518538373e-05,
      "loss": 0.0118,
      "step": 3520
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.8265882730484009,
      "learning_rate": 1.5924238363262754e-05,
      "loss": 0.0175,
      "step": 3525
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.8301231265068054,
      "learning_rate": 1.5915304207987135e-05,
      "loss": 0.0396,
      "step": 3530
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.45745956897735596,
      "learning_rate": 1.5906370052711516e-05,
      "loss": 0.0195,
      "step": 3535
    },
    {
      "epoch": 0.28,
      "grad_norm": 1.464941382408142,
      "learning_rate": 1.5897435897435897e-05,
      "loss": 0.0207,
      "step": 3540
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.7965016961097717,
      "learning_rate": 1.588850174216028e-05,
      "loss": 0.0178,
      "step": 3545
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.6353152394294739,
      "learning_rate": 1.5879567586884663e-05,
      "loss": 0.019,
      "step": 3550
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.38456305861473083,
      "learning_rate": 1.587063343160904e-05,
      "loss": 0.0169,
      "step": 3555
    },
    {
      "epoch": 0.29,
      "grad_norm": 1.2627872228622437,
      "learning_rate": 1.5861699276333425e-05,
      "loss": 0.0226,
      "step": 3560
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5265514254570007,
      "learning_rate": 1.5852765121057806e-05,
      "loss": 0.0141,
      "step": 3565
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5938783288002014,
      "learning_rate": 1.5843830965782187e-05,
      "loss": 0.0178,
      "step": 3570
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.39797863364219666,
      "learning_rate": 1.5834896810506568e-05,
      "loss": 0.0123,
      "step": 3575
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.6035088300704956,
      "learning_rate": 1.582596265523095e-05,
      "loss": 0.0162,
      "step": 3580
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.34465670585632324,
      "learning_rate": 1.581702849995533e-05,
      "loss": 0.0135,
      "step": 3585
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.40939584374427795,
      "learning_rate": 1.580809434467971e-05,
      "loss": 0.0123,
      "step": 3590
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.4972923994064331,
      "learning_rate": 1.5799160189404092e-05,
      "loss": 0.0178,
      "step": 3595
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.3403087556362152,
      "learning_rate": 1.5790226034128477e-05,
      "loss": 0.0104,
      "step": 3600
    },
    {
      "epoch": 0.29,
      "eval_loss": 0.01895288936793804,
      "eval_pearson_cosine": 0.9183723675584331,
      "eval_pearson_dot": 0.9165141926322535,
      "eval_pearson_euclidean": 0.9111139441737433,
      "eval_pearson_manhattan": 0.9105498844088082,
      "eval_pearson_max": 0.9183723675584331,
      "eval_runtime": 425.5624,
      "eval_samples_per_second": 1.175,
      "eval_spearman_cosine": 0.925627350509402,
      "eval_spearman_dot": 0.9219613838455353,
      "eval_spearman_euclidean": 0.9258841515366061,
      "eval_spearman_manhattan": 0.9260336241344965,
      "eval_spearman_max": 0.9260336241344965,
      "eval_steps_per_second": 1.175,
      "step": 3600
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.2765352129936218,
      "learning_rate": 1.5781291878852854e-05,
      "loss": 0.0154,
      "step": 3605
    },
    {
      "epoch": 0.29,
      "grad_norm": 3.1084377765655518,
      "learning_rate": 1.5772357723577235e-05,
      "loss": 0.0214,
      "step": 3610
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.42401209473609924,
      "learning_rate": 1.576342356830162e-05,
      "loss": 0.0111,
      "step": 3615
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.4498884081840515,
      "learning_rate": 1.5754489413026e-05,
      "loss": 0.0144,
      "step": 3620
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.7784323692321777,
      "learning_rate": 1.574555525775038e-05,
      "loss": 0.0164,
      "step": 3625
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.792809247970581,
      "learning_rate": 1.5736621102474763e-05,
      "loss": 0.0226,
      "step": 3630
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5559731721878052,
      "learning_rate": 1.5727686947199144e-05,
      "loss": 0.0157,
      "step": 3635
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5122278928756714,
      "learning_rate": 1.5718752791923525e-05,
      "loss": 0.0114,
      "step": 3640
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.579097330570221,
      "learning_rate": 1.5709818636647906e-05,
      "loss": 0.0176,
      "step": 3645
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.5125916004180908,
      "learning_rate": 1.5700884481372287e-05,
      "loss": 0.0112,
      "step": 3650
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.6206502318382263,
      "learning_rate": 1.569195032609667e-05,
      "loss": 0.0129,
      "step": 3655
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.578705906867981,
      "learning_rate": 1.568301617082105e-05,
      "loss": 0.0133,
      "step": 3660
    },
    {
      "epoch": 0.29,
      "grad_norm": 1.273260235786438,
      "learning_rate": 1.567408201554543e-05,
      "loss": 0.0238,
      "step": 3665
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6316016912460327,
      "learning_rate": 1.5665147860269814e-05,
      "loss": 0.0107,
      "step": 3670
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6251145005226135,
      "learning_rate": 1.5656213704994195e-05,
      "loss": 0.0115,
      "step": 3675
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.48484379053115845,
      "learning_rate": 1.5647279549718573e-05,
      "loss": 0.0171,
      "step": 3680
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.4522140920162201,
      "learning_rate": 1.5638345394442958e-05,
      "loss": 0.0222,
      "step": 3685
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6267346739768982,
      "learning_rate": 1.562941123916734e-05,
      "loss": 0.0214,
      "step": 3690
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.9123010635375977,
      "learning_rate": 1.562047708389172e-05,
      "loss": 0.0214,
      "step": 3695
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.2564802169799805,
      "learning_rate": 1.56115429286161e-05,
      "loss": 0.0209,
      "step": 3700
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.4140920341014862,
      "learning_rate": 1.5602608773340482e-05,
      "loss": 0.0194,
      "step": 3705
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.2953025102615356,
      "learning_rate": 1.5593674618064863e-05,
      "loss": 0.0175,
      "step": 3710
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.19866344332695007,
      "learning_rate": 1.5584740462789244e-05,
      "loss": 0.017,
      "step": 3715
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.5271120071411133,
      "learning_rate": 1.5575806307513625e-05,
      "loss": 0.0137,
      "step": 3720
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6619897484779358,
      "learning_rate": 1.5566872152238006e-05,
      "loss": 0.0184,
      "step": 3725
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.7720787525177002,
      "learning_rate": 1.555793799696239e-05,
      "loss": 0.0198,
      "step": 3730
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.2014611959457397,
      "learning_rate": 1.5549003841686768e-05,
      "loss": 0.0176,
      "step": 3735
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.42123496532440186,
      "learning_rate": 1.554006968641115e-05,
      "loss": 0.0136,
      "step": 3740
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.8967819213867188,
      "learning_rate": 1.5531135531135533e-05,
      "loss": 0.0202,
      "step": 3745
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.4060193598270416,
      "learning_rate": 1.5522201375859914e-05,
      "loss": 0.0183,
      "step": 3750
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6353384256362915,
      "learning_rate": 1.5513267220584295e-05,
      "loss": 0.0141,
      "step": 3755
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.3105901777744293,
      "learning_rate": 1.5504333065308677e-05,
      "loss": 0.0123,
      "step": 3760
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.33913782238960266,
      "learning_rate": 1.5495398910033058e-05,
      "loss": 0.0176,
      "step": 3765
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.664908766746521,
      "learning_rate": 1.548646475475744e-05,
      "loss": 0.0135,
      "step": 3770
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.7868657112121582,
      "learning_rate": 1.547753059948182e-05,
      "loss": 0.0122,
      "step": 3775
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6869320869445801,
      "learning_rate": 1.54685964442062e-05,
      "loss": 0.0135,
      "step": 3780
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.1669894456863403,
      "learning_rate": 1.5459662288930585e-05,
      "loss": 0.024,
      "step": 3785
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.32140159606933594,
      "learning_rate": 1.5450728133654963e-05,
      "loss": 0.0109,
      "step": 3790
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5477967262268066,
      "learning_rate": 1.5441793978379344e-05,
      "loss": 0.0132,
      "step": 3795
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.2863134741783142,
      "learning_rate": 1.5432859823103728e-05,
      "loss": 0.0225,
      "step": 3800
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.7895782589912415,
      "learning_rate": 1.542392566782811e-05,
      "loss": 0.0265,
      "step": 3805
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.36137330532073975,
      "learning_rate": 1.5414991512552487e-05,
      "loss": 0.0114,
      "step": 3810
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5505108833312988,
      "learning_rate": 1.540605735727687e-05,
      "loss": 0.0244,
      "step": 3815
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.2253755331039429,
      "learning_rate": 1.5397123202001252e-05,
      "loss": 0.0253,
      "step": 3820
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.25381264090538025,
      "learning_rate": 1.5388189046725633e-05,
      "loss": 0.0215,
      "step": 3825
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.59675133228302,
      "learning_rate": 1.5379254891450014e-05,
      "loss": 0.0229,
      "step": 3830
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.4737885296344757,
      "learning_rate": 1.5370320736174395e-05,
      "loss": 0.0172,
      "step": 3835
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.7271224856376648,
      "learning_rate": 1.536138658089878e-05,
      "loss": 0.0155,
      "step": 3840
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.3435806930065155,
      "learning_rate": 1.5352452425623158e-05,
      "loss": 0.0115,
      "step": 3845
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.37383875250816345,
      "learning_rate": 1.534351827034754e-05,
      "loss": 0.0175,
      "step": 3850
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.2769816219806671,
      "learning_rate": 1.5334584115071923e-05,
      "loss": 0.012,
      "step": 3855
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.421824187040329,
      "learning_rate": 1.5325649959796304e-05,
      "loss": 0.0168,
      "step": 3860
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.48985883593559265,
      "learning_rate": 1.5316715804520682e-05,
      "loss": 0.0121,
      "step": 3865
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.4765077829360962,
      "learning_rate": 1.5307781649245066e-05,
      "loss": 0.0181,
      "step": 3870
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.48202696442604065,
      "learning_rate": 1.5298847493969447e-05,
      "loss": 0.0157,
      "step": 3875
    },
    {
      "epoch": 0.31,
      "grad_norm": 2.033970594406128,
      "learning_rate": 1.5289913338693828e-05,
      "loss": 0.0136,
      "step": 3880
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.33205658197402954,
      "learning_rate": 1.528097918341821e-05,
      "loss": 0.012,
      "step": 3885
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.4964057207107544,
      "learning_rate": 1.527204502814259e-05,
      "loss": 0.0109,
      "step": 3890
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.9996275901794434,
      "learning_rate": 1.526311087286697e-05,
      "loss": 0.0153,
      "step": 3895
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5008103251457214,
      "learning_rate": 1.5254176717591354e-05,
      "loss": 0.0194,
      "step": 3900
    },
    {
      "epoch": 0.31,
      "eval_loss": 0.019112512469291687,
      "eval_pearson_cosine": 0.9187946359889192,
      "eval_pearson_dot": 0.916935359224543,
      "eval_pearson_euclidean": 0.9078268066209557,
      "eval_pearson_manhattan": 0.9068373107836385,
      "eval_pearson_max": 0.9187946359889192,
      "eval_runtime": 426.0704,
      "eval_samples_per_second": 1.174,
      "eval_spearman_cosine": 0.9320740642962572,
      "eval_spearman_dot": 0.9297467269869079,
      "eval_spearman_euclidean": 0.9318172632690529,
      "eval_spearman_manhattan": 0.9317702230808924,
      "eval_spearman_max": 0.9320740642962572,
      "eval_steps_per_second": 1.174,
      "step": 3900
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.5863137245178223,
      "learning_rate": 1.5245242562315733e-05,
      "loss": 0.0155,
      "step": 3905
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.9501858353614807,
      "learning_rate": 1.5236308407040114e-05,
      "loss": 0.023,
      "step": 3910
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.0070725679397583,
      "learning_rate": 1.5227374251764497e-05,
      "loss": 0.0159,
      "step": 3915
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.1535245180130005,
      "learning_rate": 1.5218440096488878e-05,
      "loss": 0.0271,
      "step": 3920
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.3500936031341553,
      "learning_rate": 1.5209505941213261e-05,
      "loss": 0.0185,
      "step": 3925
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5063101649284363,
      "learning_rate": 1.5200571785937642e-05,
      "loss": 0.0092,
      "step": 3930
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.48119908571243286,
      "learning_rate": 1.5191637630662021e-05,
      "loss": 0.0156,
      "step": 3935
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.7710624933242798,
      "learning_rate": 1.5182703475386404e-05,
      "loss": 0.0182,
      "step": 3940
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5304814577102661,
      "learning_rate": 1.5173769320110785e-05,
      "loss": 0.0127,
      "step": 3945
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5894374251365662,
      "learning_rate": 1.5164835164835166e-05,
      "loss": 0.012,
      "step": 3950
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.48509135842323303,
      "learning_rate": 1.5155901009559549e-05,
      "loss": 0.0164,
      "step": 3955
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.00001060962677,
      "learning_rate": 1.5146966854283928e-05,
      "loss": 0.0105,
      "step": 3960
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5003077983856201,
      "learning_rate": 1.513803269900831e-05,
      "loss": 0.0191,
      "step": 3965
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.4220547378063202,
      "learning_rate": 1.5129098543732692e-05,
      "loss": 0.0182,
      "step": 3970
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.353935182094574,
      "learning_rate": 1.5120164388457073e-05,
      "loss": 0.0105,
      "step": 3975
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.7862482070922852,
      "learning_rate": 1.5111230233181452e-05,
      "loss": 0.0153,
      "step": 3980
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.4707240164279938,
      "learning_rate": 1.5102296077905835e-05,
      "loss": 0.0106,
      "step": 3985
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5706127882003784,
      "learning_rate": 1.5093361922630216e-05,
      "loss": 0.0193,
      "step": 3990
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5268707275390625,
      "learning_rate": 1.5084427767354597e-05,
      "loss": 0.0115,
      "step": 3995
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.42871159315109253,
      "learning_rate": 1.507549361207898e-05,
      "loss": 0.0111,
      "step": 4000
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.47120198607444763,
      "learning_rate": 1.5066559456803361e-05,
      "loss": 0.0148,
      "step": 4005
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.6913847327232361,
      "learning_rate": 1.505762530152774e-05,
      "loss": 0.0156,
      "step": 4010
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.40243229269981384,
      "learning_rate": 1.5048691146252123e-05,
      "loss": 0.021,
      "step": 4015
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.4630833566188812,
      "learning_rate": 1.5039756990976504e-05,
      "loss": 0.0142,
      "step": 4020
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.7278119325637817,
      "learning_rate": 1.5030822835700887e-05,
      "loss": 0.0221,
      "step": 4025
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5856547951698303,
      "learning_rate": 1.5021888680425268e-05,
      "loss": 0.0127,
      "step": 4030
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.3612341582775116,
      "learning_rate": 1.5012954525149647e-05,
      "loss": 0.0142,
      "step": 4035
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.3373230993747711,
      "learning_rate": 1.500402036987403e-05,
      "loss": 0.013,
      "step": 4040
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.3555310070514679,
      "learning_rate": 1.4995086214598411e-05,
      "loss": 0.0188,
      "step": 4045
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.27012017369270325,
      "learning_rate": 1.4986152059322792e-05,
      "loss": 0.016,
      "step": 4050
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.7879682183265686,
      "learning_rate": 1.4977217904047175e-05,
      "loss": 0.019,
      "step": 4055
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.4750874638557434,
      "learning_rate": 1.4968283748771554e-05,
      "loss": 0.0156,
      "step": 4060
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5031574368476868,
      "learning_rate": 1.4959349593495935e-05,
      "loss": 0.0206,
      "step": 4065
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5057447552680969,
      "learning_rate": 1.4950415438220318e-05,
      "loss": 0.0144,
      "step": 4070
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.0315029621124268,
      "learning_rate": 1.4941481282944699e-05,
      "loss": 0.0179,
      "step": 4075
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5417433977127075,
      "learning_rate": 1.493254712766908e-05,
      "loss": 0.0126,
      "step": 4080
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.667726993560791,
      "learning_rate": 1.4923612972393463e-05,
      "loss": 0.0143,
      "step": 4085
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.8723002672195435,
      "learning_rate": 1.4914678817117842e-05,
      "loss": 0.012,
      "step": 4090
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.8944787979125977,
      "learning_rate": 1.4905744661842223e-05,
      "loss": 0.0205,
      "step": 4095
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5145695805549622,
      "learning_rate": 1.4896810506566606e-05,
      "loss": 0.0118,
      "step": 4100
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.9632664322853088,
      "learning_rate": 1.4887876351290987e-05,
      "loss": 0.0233,
      "step": 4105
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.4608353078365326,
      "learning_rate": 1.487894219601537e-05,
      "loss": 0.0192,
      "step": 4110
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.8873416781425476,
      "learning_rate": 1.4870008040739749e-05,
      "loss": 0.0196,
      "step": 4115
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.7420912384986877,
      "learning_rate": 1.486107388546413e-05,
      "loss": 0.0114,
      "step": 4120
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.3144592344760895,
      "learning_rate": 1.4852139730188513e-05,
      "loss": 0.0136,
      "step": 4125
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.4143444001674652,
      "learning_rate": 1.4843205574912894e-05,
      "loss": 0.0102,
      "step": 4130
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.3450114727020264,
      "learning_rate": 1.4834271419637275e-05,
      "loss": 0.0273,
      "step": 4135
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.6474159955978394,
      "learning_rate": 1.4825337264361657e-05,
      "loss": 0.0125,
      "step": 4140
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.3440127372741699,
      "learning_rate": 1.4816403109086037e-05,
      "loss": 0.016,
      "step": 4145
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.6236218214035034,
      "learning_rate": 1.4807468953810418e-05,
      "loss": 0.019,
      "step": 4150
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.9320889115333557,
      "learning_rate": 1.47985347985348e-05,
      "loss": 0.015,
      "step": 4155
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.36292609572410583,
      "learning_rate": 1.4789600643259182e-05,
      "loss": 0.0155,
      "step": 4160
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.278648018836975,
      "learning_rate": 1.4780666487983561e-05,
      "loss": 0.0229,
      "step": 4165
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.24334418773651123,
      "learning_rate": 1.4771732332707944e-05,
      "loss": 0.0099,
      "step": 4170
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.7839992046356201,
      "learning_rate": 1.4762798177432325e-05,
      "loss": 0.0167,
      "step": 4175
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.3292075395584106,
      "learning_rate": 1.4753864022156706e-05,
      "loss": 0.0168,
      "step": 4180
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.4192187488079071,
      "learning_rate": 1.4744929866881089e-05,
      "loss": 0.0177,
      "step": 4185
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5623775124549866,
      "learning_rate": 1.4735995711605468e-05,
      "loss": 0.0103,
      "step": 4190
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5713854432106018,
      "learning_rate": 1.4727061556329849e-05,
      "loss": 0.011,
      "step": 4195
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.6574892997741699,
      "learning_rate": 1.4718127401054232e-05,
      "loss": 0.0169,
      "step": 4200
    },
    {
      "epoch": 0.34,
      "eval_loss": 0.01563582755625248,
      "eval_pearson_cosine": 0.93071357698338,
      "eval_pearson_dot": 0.9303618828257327,
      "eval_pearson_euclidean": 0.9158033535152222,
      "eval_pearson_manhattan": 0.9136122746167228,
      "eval_pearson_max": 0.93071357698338,
      "eval_runtime": 426.316,
      "eval_samples_per_second": 1.173,
      "eval_spearman_cosine": 0.9357050868203473,
      "eval_spearman_dot": 0.9326605306421225,
      "eval_spearman_euclidean": 0.9346635786543147,
      "eval_spearman_manhattan": 0.9334011736046943,
      "eval_spearman_max": 0.9357050868203473,
      "eval_steps_per_second": 1.173,
      "step": 4200
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5762566328048706,
      "learning_rate": 1.4709193245778613e-05,
      "loss": 0.0296,
      "step": 4205
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.6294378638267517,
      "learning_rate": 1.4700259090502995e-05,
      "loss": 0.0158,
      "step": 4210
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5651215314865112,
      "learning_rate": 1.4691324935227376e-05,
      "loss": 0.0131,
      "step": 4215
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.6574472784996033,
      "learning_rate": 1.4682390779951756e-05,
      "loss": 0.0166,
      "step": 4220
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.4540180563926697,
      "learning_rate": 1.4673456624676139e-05,
      "loss": 0.024,
      "step": 4225
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.3493760824203491,
      "learning_rate": 1.466452246940052e-05,
      "loss": 0.0149,
      "step": 4230
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.0167644023895264,
      "learning_rate": 1.46555883141249e-05,
      "loss": 0.0149,
      "step": 4235
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.9241525530815125,
      "learning_rate": 1.4646654158849283e-05,
      "loss": 0.0184,
      "step": 4240
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.6745620965957642,
      "learning_rate": 1.4637720003573663e-05,
      "loss": 0.0158,
      "step": 4245
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.4359780251979828,
      "learning_rate": 1.4628785848298044e-05,
      "loss": 0.0098,
      "step": 4250
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.4030389189720154,
      "learning_rate": 1.4619851693022426e-05,
      "loss": 0.0184,
      "step": 4255
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5261355638504028,
      "learning_rate": 1.4610917537746807e-05,
      "loss": 0.0141,
      "step": 4260
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.38097327947616577,
      "learning_rate": 1.4601983382471187e-05,
      "loss": 0.0083,
      "step": 4265
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.41486430168151855,
      "learning_rate": 1.459304922719557e-05,
      "loss": 0.0202,
      "step": 4270
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.5645167827606201,
      "learning_rate": 1.458411507191995e-05,
      "loss": 0.0126,
      "step": 4275
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.4771600365638733,
      "learning_rate": 1.4575180916644332e-05,
      "loss": 0.0103,
      "step": 4280
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.8267874717712402,
      "learning_rate": 1.4566246761368714e-05,
      "loss": 0.0182,
      "step": 4285
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.7216476202011108,
      "learning_rate": 1.4557312606093095e-05,
      "loss": 0.0155,
      "step": 4290
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.0095362663269043,
      "learning_rate": 1.4548378450817478e-05,
      "loss": 0.0199,
      "step": 4295
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.36790329217910767,
      "learning_rate": 1.4539444295541857e-05,
      "loss": 0.0231,
      "step": 4300
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.4778492748737335,
      "learning_rate": 1.4530510140266239e-05,
      "loss": 0.0101,
      "step": 4305
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.3736858665943146,
      "learning_rate": 1.4521575984990621e-05,
      "loss": 0.0157,
      "step": 4310
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5705660581588745,
      "learning_rate": 1.4512641829715002e-05,
      "loss": 0.0133,
      "step": 4315
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.7104588747024536,
      "learning_rate": 1.4503707674439382e-05,
      "loss": 0.0209,
      "step": 4320
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.4687543511390686,
      "learning_rate": 1.4494773519163764e-05,
      "loss": 0.0072,
      "step": 4325
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5142909288406372,
      "learning_rate": 1.4485839363888145e-05,
      "loss": 0.0127,
      "step": 4330
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.3397541046142578,
      "learning_rate": 1.4476905208612526e-05,
      "loss": 0.0122,
      "step": 4335
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.4727453589439392,
      "learning_rate": 1.446797105333691e-05,
      "loss": 0.0212,
      "step": 4340
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.7717282176017761,
      "learning_rate": 1.445903689806129e-05,
      "loss": 0.0277,
      "step": 4345
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.4375796914100647,
      "learning_rate": 1.445010274278567e-05,
      "loss": 0.0119,
      "step": 4350
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.45401817560195923,
      "learning_rate": 1.4441168587510052e-05,
      "loss": 0.0128,
      "step": 4355
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.4388526976108551,
      "learning_rate": 1.4432234432234433e-05,
      "loss": 0.0181,
      "step": 4360
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.4506477415561676,
      "learning_rate": 1.4423300276958814e-05,
      "loss": 0.0118,
      "step": 4365
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.7086319923400879,
      "learning_rate": 1.4414366121683197e-05,
      "loss": 0.0111,
      "step": 4370
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.185771182179451,
      "learning_rate": 1.4405431966407576e-05,
      "loss": 0.0151,
      "step": 4375
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.26904481649398804,
      "learning_rate": 1.4396497811131958e-05,
      "loss": 0.0116,
      "step": 4380
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.8145022392272949,
      "learning_rate": 1.438756365585634e-05,
      "loss": 0.0097,
      "step": 4385
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.1931172609329224,
      "learning_rate": 1.4378629500580721e-05,
      "loss": 0.0158,
      "step": 4390
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.37602362036705017,
      "learning_rate": 1.4369695345305104e-05,
      "loss": 0.0145,
      "step": 4395
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5753958225250244,
      "learning_rate": 1.4360761190029483e-05,
      "loss": 0.0156,
      "step": 4400
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.6079641580581665,
      "learning_rate": 1.4351827034753864e-05,
      "loss": 0.0132,
      "step": 4405
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.35592758655548096,
      "learning_rate": 1.4342892879478247e-05,
      "loss": 0.0152,
      "step": 4410
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.3908535838127136,
      "learning_rate": 1.4333958724202628e-05,
      "loss": 0.0234,
      "step": 4415
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.4124375581741333,
      "learning_rate": 1.432502456892701e-05,
      "loss": 0.0165,
      "step": 4420
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.31878653168678284,
      "learning_rate": 1.4316090413651392e-05,
      "loss": 0.0134,
      "step": 4425
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.6826820373535156,
      "learning_rate": 1.4307156258375771e-05,
      "loss": 0.0108,
      "step": 4430
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5857677459716797,
      "learning_rate": 1.4298222103100152e-05,
      "loss": 0.01,
      "step": 4435
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.7342239618301392,
      "learning_rate": 1.4289287947824535e-05,
      "loss": 0.018,
      "step": 4440
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.6974345445632935,
      "learning_rate": 1.4280353792548916e-05,
      "loss": 0.0103,
      "step": 4445
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5381551384925842,
      "learning_rate": 1.4271419637273295e-05,
      "loss": 0.0107,
      "step": 4450
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.46334195137023926,
      "learning_rate": 1.4262485481997678e-05,
      "loss": 0.0087,
      "step": 4455
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.9317929744720459,
      "learning_rate": 1.425355132672206e-05,
      "loss": 0.0167,
      "step": 4460
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5066403746604919,
      "learning_rate": 1.424461717144644e-05,
      "loss": 0.0169,
      "step": 4465
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.6555119156837463,
      "learning_rate": 1.4235683016170823e-05,
      "loss": 0.017,
      "step": 4470
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.7735246419906616,
      "learning_rate": 1.4226748860895202e-05,
      "loss": 0.0136,
      "step": 4475
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.6732664108276367,
      "learning_rate": 1.4217814705619585e-05,
      "loss": 0.0157,
      "step": 4480
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.6014764904975891,
      "learning_rate": 1.4208880550343966e-05,
      "loss": 0.0136,
      "step": 4485
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.6695230007171631,
      "learning_rate": 1.4199946395068347e-05,
      "loss": 0.0128,
      "step": 4490
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.4343644380569458,
      "learning_rate": 1.419101223979273e-05,
      "loss": 0.0095,
      "step": 4495
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.8279352784156799,
      "learning_rate": 1.4182078084517111e-05,
      "loss": 0.0115,
      "step": 4500
    },
    {
      "epoch": 0.36,
      "eval_loss": 0.015759816393256187,
      "eval_pearson_cosine": 0.9315854318304321,
      "eval_pearson_dot": 0.9286307914356223,
      "eval_pearson_euclidean": 0.919714470249928,
      "eval_pearson_manhattan": 0.9173628838845561,
      "eval_pearson_max": 0.9315854318304321,
      "eval_runtime": 425.7927,
      "eval_samples_per_second": 1.174,
      "eval_spearman_cosine": 0.9374404857619429,
      "eval_spearman_dot": 0.9333967575870303,
      "eval_spearman_euclidean": 0.9374683258733034,
      "eval_spearman_manhattan": 0.9356461425845704,
      "eval_spearman_max": 0.9374683258733034,
      "eval_steps_per_second": 1.174,
      "step": 4500
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5468992590904236,
      "learning_rate": 1.417314392924149e-05,
      "loss": 0.0102,
      "step": 4505
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.4848220646381378,
      "learning_rate": 1.4164209773965873e-05,
      "loss": 0.0104,
      "step": 4510
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.43585026264190674,
      "learning_rate": 1.4155275618690254e-05,
      "loss": 0.0112,
      "step": 4515
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5157576203346252,
      "learning_rate": 1.4146341463414635e-05,
      "loss": 0.0138,
      "step": 4520
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.44182753562927246,
      "learning_rate": 1.4137407308139018e-05,
      "loss": 0.0141,
      "step": 4525
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.6160708665847778,
      "learning_rate": 1.4128473152863397e-05,
      "loss": 0.0124,
      "step": 4530
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.3861226439476013,
      "learning_rate": 1.4119538997587778e-05,
      "loss": 0.0142,
      "step": 4535
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.1385138034820557,
      "learning_rate": 1.4110604842312161e-05,
      "loss": 0.0112,
      "step": 4540
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.8177729249000549,
      "learning_rate": 1.4101670687036542e-05,
      "loss": 0.0196,
      "step": 4545
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.8417105078697205,
      "learning_rate": 1.4092736531760923e-05,
      "loss": 0.0178,
      "step": 4550
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5690799355506897,
      "learning_rate": 1.4083802376485306e-05,
      "loss": 0.0126,
      "step": 4555
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5417916178703308,
      "learning_rate": 1.4074868221209685e-05,
      "loss": 0.0248,
      "step": 4560
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.2451568841934204,
      "learning_rate": 1.4065934065934068e-05,
      "loss": 0.0256,
      "step": 4565
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.6899533271789551,
      "learning_rate": 1.4056999910658449e-05,
      "loss": 0.0116,
      "step": 4570
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.9121742844581604,
      "learning_rate": 1.404806575538283e-05,
      "loss": 0.0211,
      "step": 4575
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.6731112599372864,
      "learning_rate": 1.4039131600107213e-05,
      "loss": 0.0172,
      "step": 4580
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3643783628940582,
      "learning_rate": 1.4030197444831592e-05,
      "loss": 0.0143,
      "step": 4585
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.1846754550933838,
      "learning_rate": 1.4021263289555973e-05,
      "loss": 0.011,
      "step": 4590
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.2550729513168335,
      "learning_rate": 1.4012329134280356e-05,
      "loss": 0.0116,
      "step": 4595
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3365246653556824,
      "learning_rate": 1.4003394979004737e-05,
      "loss": 0.0087,
      "step": 4600
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.699928343296051,
      "learning_rate": 1.3994460823729116e-05,
      "loss": 0.0121,
      "step": 4605
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.3738575577735901,
      "learning_rate": 1.3985526668453499e-05,
      "loss": 0.0122,
      "step": 4610
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5205432772636414,
      "learning_rate": 1.397659251317788e-05,
      "loss": 0.0102,
      "step": 4615
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.0370134115219116,
      "learning_rate": 1.3967658357902261e-05,
      "loss": 0.0154,
      "step": 4620
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.8849940299987793,
      "learning_rate": 1.3958724202626644e-05,
      "loss": 0.0183,
      "step": 4625
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.4405355155467987,
      "learning_rate": 1.3949790047351025e-05,
      "loss": 0.0146,
      "step": 4630
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.4738609194755554,
      "learning_rate": 1.3940855892075404e-05,
      "loss": 0.0099,
      "step": 4635
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5049815773963928,
      "learning_rate": 1.3931921736799787e-05,
      "loss": 0.0092,
      "step": 4640
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.2640978991985321,
      "learning_rate": 1.3922987581524168e-05,
      "loss": 0.0094,
      "step": 4645
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.6783714294433594,
      "learning_rate": 1.3914053426248549e-05,
      "loss": 0.0112,
      "step": 4650
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.9599121809005737,
      "learning_rate": 1.3905119270972932e-05,
      "loss": 0.0135,
      "step": 4655
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.2599102556705475,
      "learning_rate": 1.3896185115697311e-05,
      "loss": 0.0087,
      "step": 4660
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5762407779693604,
      "learning_rate": 1.3887250960421694e-05,
      "loss": 0.0122,
      "step": 4665
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.9780054688453674,
      "learning_rate": 1.3878316805146075e-05,
      "loss": 0.0205,
      "step": 4670
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.742286741733551,
      "learning_rate": 1.3869382649870456e-05,
      "loss": 0.0149,
      "step": 4675
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.828479528427124,
      "learning_rate": 1.3860448494594838e-05,
      "loss": 0.0166,
      "step": 4680
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.460374116897583,
      "learning_rate": 1.3851514339319218e-05,
      "loss": 0.0266,
      "step": 4685
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.4525332748889923,
      "learning_rate": 1.3842580184043599e-05,
      "loss": 0.0133,
      "step": 4690
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.40976735949516296,
      "learning_rate": 1.3833646028767982e-05,
      "loss": 0.0119,
      "step": 4695
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.30218005180358887,
      "learning_rate": 1.3824711873492363e-05,
      "loss": 0.0186,
      "step": 4700
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.3435049057006836,
      "learning_rate": 1.3815777718216744e-05,
      "loss": 0.0235,
      "step": 4705
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.19714932143688202,
      "learning_rate": 1.3806843562941126e-05,
      "loss": 0.0114,
      "step": 4710
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.3355713188648224,
      "learning_rate": 1.3797909407665506e-05,
      "loss": 0.0182,
      "step": 4715
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.8462282419204712,
      "learning_rate": 1.3788975252389887e-05,
      "loss": 0.0161,
      "step": 4720
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.3174177408218384,
      "learning_rate": 1.378004109711427e-05,
      "loss": 0.0092,
      "step": 4725
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5798084139823914,
      "learning_rate": 1.377110694183865e-05,
      "loss": 0.0159,
      "step": 4730
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.7595842480659485,
      "learning_rate": 1.376217278656303e-05,
      "loss": 0.019,
      "step": 4735
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5525384545326233,
      "learning_rate": 1.3753238631287413e-05,
      "loss": 0.0122,
      "step": 4740
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.7659673690795898,
      "learning_rate": 1.3744304476011794e-05,
      "loss": 0.0184,
      "step": 4745
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.7609495520591736,
      "learning_rate": 1.3735370320736176e-05,
      "loss": 0.0114,
      "step": 4750
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.5531354546546936,
      "learning_rate": 1.3726436165460557e-05,
      "loss": 0.01,
      "step": 4755
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.6335747241973877,
      "learning_rate": 1.3717502010184938e-05,
      "loss": 0.0171,
      "step": 4760
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.3479459583759308,
      "learning_rate": 1.3708567854909321e-05,
      "loss": 0.012,
      "step": 4765
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.31756049394607544,
      "learning_rate": 1.36996336996337e-05,
      "loss": 0.0101,
      "step": 4770
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.2414575815200806,
      "learning_rate": 1.3690699544358082e-05,
      "loss": 0.0171,
      "step": 4775
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.1781928390264511,
      "learning_rate": 1.3681765389082464e-05,
      "loss": 0.0189,
      "step": 4780
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.4548472464084625,
      "learning_rate": 1.3672831233806845e-05,
      "loss": 0.0089,
      "step": 4785
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.3812129497528076,
      "learning_rate": 1.3663897078531225e-05,
      "loss": 0.0115,
      "step": 4790
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.44192445278167725,
      "learning_rate": 1.3654962923255607e-05,
      "loss": 0.0119,
      "step": 4795
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.4043430089950562,
      "learning_rate": 1.3646028767979988e-05,
      "loss": 0.02,
      "step": 4800
    },
    {
      "epoch": 0.39,
      "eval_loss": 0.017762023955583572,
      "eval_pearson_cosine": 0.9247524125145,
      "eval_pearson_dot": 0.9227442447856763,
      "eval_pearson_euclidean": 0.9166059342093679,
      "eval_pearson_manhattan": 0.9150588699348506,
      "eval_pearson_max": 0.9247524125145,
      "eval_runtime": 426.6962,
      "eval_samples_per_second": 1.172,
      "eval_spearman_cosine": 0.9294770619082475,
      "eval_spearman_dot": 0.9266574346297385,
      "eval_spearman_euclidean": 0.9284289297157189,
      "eval_spearman_manhattan": 0.927501086004344,
      "eval_spearman_max": 0.9294770619082475,
      "eval_steps_per_second": 1.172,
      "step": 4800
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.41596701741218567,
      "learning_rate": 1.363709461270437e-05,
      "loss": 0.0173,
      "step": 4805
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.505337119102478,
      "learning_rate": 1.3628160457428752e-05,
      "loss": 0.0075,
      "step": 4810
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.4152344763278961,
      "learning_rate": 1.3619226302153132e-05,
      "loss": 0.0136,
      "step": 4815
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.9099875092506409,
      "learning_rate": 1.3610292146877513e-05,
      "loss": 0.0118,
      "step": 4820
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.6013542413711548,
      "learning_rate": 1.3601357991601895e-05,
      "loss": 0.014,
      "step": 4825
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.409640371799469,
      "learning_rate": 1.3592423836326276e-05,
      "loss": 0.015,
      "step": 4830
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.3172929286956787,
      "learning_rate": 1.3583489681050657e-05,
      "loss": 0.017,
      "step": 4835
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5413698554039001,
      "learning_rate": 1.357455552577504e-05,
      "loss": 0.0137,
      "step": 4840
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.6152754426002502,
      "learning_rate": 1.356562137049942e-05,
      "loss": 0.0127,
      "step": 4845
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5605708360671997,
      "learning_rate": 1.3556687215223802e-05,
      "loss": 0.0155,
      "step": 4850
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.4922347068786621,
      "learning_rate": 1.3547753059948183e-05,
      "loss": 0.0125,
      "step": 4855
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.2988830506801605,
      "learning_rate": 1.3538818904672564e-05,
      "loss": 0.0117,
      "step": 4860
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.837377667427063,
      "learning_rate": 1.3529884749396947e-05,
      "loss": 0.0172,
      "step": 4865
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.420107364654541,
      "learning_rate": 1.3520950594121326e-05,
      "loss": 0.0119,
      "step": 4870
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.4992135763168335,
      "learning_rate": 1.3512016438845707e-05,
      "loss": 0.0092,
      "step": 4875
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.7514760494232178,
      "learning_rate": 1.350308228357009e-05,
      "loss": 0.0248,
      "step": 4880
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5291612148284912,
      "learning_rate": 1.3494148128294471e-05,
      "loss": 0.016,
      "step": 4885
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.44455814361572266,
      "learning_rate": 1.348521397301885e-05,
      "loss": 0.0145,
      "step": 4890
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.39508140087127686,
      "learning_rate": 1.3476279817743233e-05,
      "loss": 0.014,
      "step": 4895
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.5380381941795349,
      "learning_rate": 1.3467345662467614e-05,
      "loss": 0.0168,
      "step": 4900
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.16052573919296265,
      "learning_rate": 1.3458411507191995e-05,
      "loss": 0.0087,
      "step": 4905
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.33839181065559387,
      "learning_rate": 1.3449477351916378e-05,
      "loss": 0.0144,
      "step": 4910
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.43585050106048584,
      "learning_rate": 1.3440543196640759e-05,
      "loss": 0.0131,
      "step": 4915
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.20657236874103546,
      "learning_rate": 1.3431609041365138e-05,
      "loss": 0.013,
      "step": 4920
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.38462182879447937,
      "learning_rate": 1.3422674886089521e-05,
      "loss": 0.0094,
      "step": 4925
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.6109427213668823,
      "learning_rate": 1.3413740730813902e-05,
      "loss": 0.0161,
      "step": 4930
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.39142367243766785,
      "learning_rate": 1.3404806575538285e-05,
      "loss": 0.0146,
      "step": 4935
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.8469399213790894,
      "learning_rate": 1.3395872420262666e-05,
      "loss": 0.0136,
      "step": 4940
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.40543830394744873,
      "learning_rate": 1.3386938264987045e-05,
      "loss": 0.0122,
      "step": 4945
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3752301335334778,
      "learning_rate": 1.3378004109711428e-05,
      "loss": 0.0121,
      "step": 4950
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3208424150943756,
      "learning_rate": 1.3369069954435809e-05,
      "loss": 0.0117,
      "step": 4955
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.2412027269601822,
      "learning_rate": 1.336013579916019e-05,
      "loss": 0.012,
      "step": 4960
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.4787154495716095,
      "learning_rate": 1.3351201643884573e-05,
      "loss": 0.0116,
      "step": 4965
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3224911689758301,
      "learning_rate": 1.3342267488608954e-05,
      "loss": 0.0088,
      "step": 4970
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.46356651186943054,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.0127,
      "step": 4975
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5258038640022278,
      "learning_rate": 1.3324399178057716e-05,
      "loss": 0.0171,
      "step": 4980
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.6068939566612244,
      "learning_rate": 1.3315465022782097e-05,
      "loss": 0.0125,
      "step": 4985
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3395962715148926,
      "learning_rate": 1.3306530867506478e-05,
      "loss": 0.0103,
      "step": 4990
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.4415256679058075,
      "learning_rate": 1.329759671223086e-05,
      "loss": 0.0109,
      "step": 4995
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.4051103889942169,
      "learning_rate": 1.328866255695524e-05,
      "loss": 0.0148,
      "step": 5000
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.085768461227417,
      "learning_rate": 1.3279728401679621e-05,
      "loss": 0.0168,
      "step": 5005
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.38266754150390625,
      "learning_rate": 1.3270794246404004e-05,
      "loss": 0.0096,
      "step": 5010
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.6186217069625854,
      "learning_rate": 1.3261860091128385e-05,
      "loss": 0.0108,
      "step": 5015
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.36592113971710205,
      "learning_rate": 1.3252925935852764e-05,
      "loss": 0.0115,
      "step": 5020
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5204969048500061,
      "learning_rate": 1.3243991780577147e-05,
      "loss": 0.0141,
      "step": 5025
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.532606840133667,
      "learning_rate": 1.3235057625301528e-05,
      "loss": 0.015,
      "step": 5030
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.755771815776825,
      "learning_rate": 1.322612347002591e-05,
      "loss": 0.0221,
      "step": 5035
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.0052992105484009,
      "learning_rate": 1.3217189314750292e-05,
      "loss": 0.0124,
      "step": 5040
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5117295384407043,
      "learning_rate": 1.3208255159474673e-05,
      "loss": 0.0168,
      "step": 5045
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.7341750860214233,
      "learning_rate": 1.3199321004199056e-05,
      "loss": 0.009,
      "step": 5050
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.4001040756702423,
      "learning_rate": 1.3190386848923435e-05,
      "loss": 0.0161,
      "step": 5055
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5844860672950745,
      "learning_rate": 1.3181452693647816e-05,
      "loss": 0.0166,
      "step": 5060
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.49384939670562744,
      "learning_rate": 1.3172518538372199e-05,
      "loss": 0.0108,
      "step": 5065
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.352339267730713,
      "learning_rate": 1.316358438309658e-05,
      "loss": 0.0101,
      "step": 5070
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5237613916397095,
      "learning_rate": 1.3154650227820959e-05,
      "loss": 0.0128,
      "step": 5075
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5380772948265076,
      "learning_rate": 1.3145716072545342e-05,
      "loss": 0.0095,
      "step": 5080
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.42861300706863403,
      "learning_rate": 1.3136781917269723e-05,
      "loss": 0.0113,
      "step": 5085
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.531801700592041,
      "learning_rate": 1.3127847761994104e-05,
      "loss": 0.0168,
      "step": 5090
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.3549353778362274,
      "learning_rate": 1.3118913606718487e-05,
      "loss": 0.0137,
      "step": 5095
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5445482730865479,
      "learning_rate": 1.3109979451442866e-05,
      "loss": 0.0116,
      "step": 5100
    },
    {
      "epoch": 0.41,
      "eval_loss": 0.014637832529842854,
      "eval_pearson_cosine": 0.9310489858999218,
      "eval_pearson_dot": 0.9284906820904276,
      "eval_pearson_euclidean": 0.9189425206931262,
      "eval_pearson_manhattan": 0.9155948344081257,
      "eval_pearson_max": 0.9310489858999218,
      "eval_runtime": 426.3308,
      "eval_samples_per_second": 1.173,
      "eval_spearman_cosine": 0.9312627650510603,
      "eval_spearman_dot": 0.9258572714290856,
      "eval_spearman_euclidean": 0.9319769119076476,
      "eval_spearman_manhattan": 0.9306522986091945,
      "eval_spearman_max": 0.9319769119076476,
      "eval_steps_per_second": 1.173,
      "step": 5100
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.9755122661590576,
      "learning_rate": 1.3101045296167247e-05,
      "loss": 0.0151,
      "step": 5105
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5714341998100281,
      "learning_rate": 1.309211114089163e-05,
      "loss": 0.0182,
      "step": 5110
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.8099054098129272,
      "learning_rate": 1.3083176985616011e-05,
      "loss": 0.0154,
      "step": 5115
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5712036490440369,
      "learning_rate": 1.3074242830340394e-05,
      "loss": 0.0132,
      "step": 5120
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.4956033527851105,
      "learning_rate": 1.3065308675064775e-05,
      "loss": 0.0146,
      "step": 5125
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.0638816356658936,
      "learning_rate": 1.3056374519789154e-05,
      "loss": 0.0201,
      "step": 5130
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.661064863204956,
      "learning_rate": 1.3047440364513537e-05,
      "loss": 0.0221,
      "step": 5135
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.4543398320674896,
      "learning_rate": 1.3038506209237918e-05,
      "loss": 0.0102,
      "step": 5140
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.422681987285614,
      "learning_rate": 1.3029572053962299e-05,
      "loss": 0.0204,
      "step": 5145
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.2622760832309723,
      "learning_rate": 1.3020637898686682e-05,
      "loss": 0.0106,
      "step": 5150
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.563801109790802,
      "learning_rate": 1.3011703743411061e-05,
      "loss": 0.0088,
      "step": 5155
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.7994465231895447,
      "learning_rate": 1.3002769588135442e-05,
      "loss": 0.0166,
      "step": 5160
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.3288549780845642,
      "learning_rate": 1.2993835432859825e-05,
      "loss": 0.0103,
      "step": 5165
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5800594091415405,
      "learning_rate": 1.2984901277584206e-05,
      "loss": 0.0159,
      "step": 5170
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.37298092246055603,
      "learning_rate": 1.2975967122308587e-05,
      "loss": 0.0104,
      "step": 5175
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.35554614663124084,
      "learning_rate": 1.296703296703297e-05,
      "loss": 0.0107,
      "step": 5180
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.20213966071605682,
      "learning_rate": 1.2958098811757349e-05,
      "loss": 0.0098,
      "step": 5185
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.2717779576778412,
      "learning_rate": 1.294916465648173e-05,
      "loss": 0.0112,
      "step": 5190
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.4269130527973175,
      "learning_rate": 1.2940230501206113e-05,
      "loss": 0.0169,
      "step": 5195
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.9823828935623169,
      "learning_rate": 1.2931296345930494e-05,
      "loss": 0.0167,
      "step": 5200
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.52610844373703,
      "learning_rate": 1.2922362190654876e-05,
      "loss": 0.0135,
      "step": 5205
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.8198827505111694,
      "learning_rate": 1.2913428035379256e-05,
      "loss": 0.0192,
      "step": 5210
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.587947428226471,
      "learning_rate": 1.2904493880103637e-05,
      "loss": 0.0122,
      "step": 5215
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.8936262726783752,
      "learning_rate": 1.289555972482802e-05,
      "loss": 0.0159,
      "step": 5220
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.42766860127449036,
      "learning_rate": 1.28866255695524e-05,
      "loss": 0.0249,
      "step": 5225
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.59279465675354,
      "learning_rate": 1.287769141427678e-05,
      "loss": 0.0125,
      "step": 5230
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.9216848015785217,
      "learning_rate": 1.2868757259001163e-05,
      "loss": 0.0138,
      "step": 5235
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.30807581543922424,
      "learning_rate": 1.2859823103725544e-05,
      "loss": 0.0082,
      "step": 5240
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.7523232102394104,
      "learning_rate": 1.2850888948449925e-05,
      "loss": 0.0183,
      "step": 5245
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.29670125246047974,
      "learning_rate": 1.2841954793174307e-05,
      "loss": 0.0121,
      "step": 5250
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.43117162585258484,
      "learning_rate": 1.2833020637898688e-05,
      "loss": 0.0111,
      "step": 5255
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.49894264340400696,
      "learning_rate": 1.2824086482623068e-05,
      "loss": 0.015,
      "step": 5260
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.3253922760486603,
      "learning_rate": 1.281515232734745e-05,
      "loss": 0.0104,
      "step": 5265
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.47245854139328003,
      "learning_rate": 1.2806218172071832e-05,
      "loss": 0.0083,
      "step": 5270
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.3503600060939789,
      "learning_rate": 1.2797284016796213e-05,
      "loss": 0.0098,
      "step": 5275
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.4951741099357605,
      "learning_rate": 1.2788349861520595e-05,
      "loss": 0.0137,
      "step": 5280
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.7028777599334717,
      "learning_rate": 1.2779415706244975e-05,
      "loss": 0.0255,
      "step": 5285
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.7068243026733398,
      "learning_rate": 1.2770481550969356e-05,
      "loss": 0.0115,
      "step": 5290
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.3059793710708618,
      "learning_rate": 1.2761547395693738e-05,
      "loss": 0.0117,
      "step": 5295
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.9232611060142517,
      "learning_rate": 1.275261324041812e-05,
      "loss": 0.0174,
      "step": 5300
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.536094069480896,
      "learning_rate": 1.2743679085142502e-05,
      "loss": 0.0105,
      "step": 5305
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.3886105716228485,
      "learning_rate": 1.2734744929866883e-05,
      "loss": 0.0087,
      "step": 5310
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5662040710449219,
      "learning_rate": 1.2725810774591263e-05,
      "loss": 0.0308,
      "step": 5315
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5094872117042542,
      "learning_rate": 1.2716876619315645e-05,
      "loss": 0.016,
      "step": 5320
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.7154682278633118,
      "learning_rate": 1.2707942464040026e-05,
      "loss": 0.0268,
      "step": 5325
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.6192654967308044,
      "learning_rate": 1.2699008308764407e-05,
      "loss": 0.0165,
      "step": 5330
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.4172917902469635,
      "learning_rate": 1.269007415348879e-05,
      "loss": 0.0088,
      "step": 5335
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.19832268357276917,
      "learning_rate": 1.268113999821317e-05,
      "loss": 0.0099,
      "step": 5340
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.2940590977668762,
      "learning_rate": 1.267220584293755e-05,
      "loss": 0.0114,
      "step": 5345
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5283774137496948,
      "learning_rate": 1.2663271687661933e-05,
      "loss": 0.0118,
      "step": 5350
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.37739717960357666,
      "learning_rate": 1.2654337532386314e-05,
      "loss": 0.0137,
      "step": 5355
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.4239113926887512,
      "learning_rate": 1.2645403377110694e-05,
      "loss": 0.0161,
      "step": 5360
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.7093974947929382,
      "learning_rate": 1.2636469221835076e-05,
      "loss": 0.0121,
      "step": 5365
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.26933395862579346,
      "learning_rate": 1.2627535066559457e-05,
      "loss": 0.0138,
      "step": 5370
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5276101231575012,
      "learning_rate": 1.2618600911283838e-05,
      "loss": 0.0123,
      "step": 5375
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.5479996800422668,
      "learning_rate": 1.2609666756008221e-05,
      "loss": 0.013,
      "step": 5380
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.2765143811702728,
      "learning_rate": 1.2600732600732602e-05,
      "loss": 0.0115,
      "step": 5385
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.2520207166671753,
      "learning_rate": 1.2591798445456985e-05,
      "loss": 0.0135,
      "step": 5390
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.5313918590545654,
      "learning_rate": 1.2582864290181364e-05,
      "loss": 0.0129,
      "step": 5395
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.37574347853660583,
      "learning_rate": 1.2573930134905745e-05,
      "loss": 0.0116,
      "step": 5400
    },
    {
      "epoch": 0.43,
      "eval_loss": 0.013623657636344433,
      "eval_pearson_cosine": 0.938319624187462,
      "eval_pearson_dot": 0.9349658820196324,
      "eval_pearson_euclidean": 0.9255554232607915,
      "eval_pearson_manhattan": 0.9221635273240761,
      "eval_pearson_max": 0.938319624187462,
      "eval_runtime": 426.6639,
      "eval_samples_per_second": 1.172,
      "eval_spearman_cosine": 0.9444973779895118,
      "eval_spearman_dot": 0.9385887463549853,
      "eval_spearman_euclidean": 0.9453539894159576,
      "eval_spearman_manhattan": 0.9434634538538154,
      "eval_spearman_max": 0.9453539894159576,
      "eval_steps_per_second": 1.172,
      "step": 5400
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.32526862621307373,
      "learning_rate": 1.2564995979630128e-05,
      "loss": 0.0071,
      "step": 5405
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.6559897065162659,
      "learning_rate": 1.2556061824354509e-05,
      "loss": 0.0134,
      "step": 5410
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.8771166801452637,
      "learning_rate": 1.2547127669078888e-05,
      "loss": 0.0195,
      "step": 5415
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.9443796873092651,
      "learning_rate": 1.2538193513803271e-05,
      "loss": 0.0146,
      "step": 5420
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.46752771735191345,
      "learning_rate": 1.2529259358527652e-05,
      "loss": 0.0119,
      "step": 5425
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.5060916543006897,
      "learning_rate": 1.2520325203252033e-05,
      "loss": 0.0181,
      "step": 5430
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.6324385404586792,
      "learning_rate": 1.2511391047976416e-05,
      "loss": 0.0252,
      "step": 5435
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.12856125831604,
      "learning_rate": 1.2502456892700795e-05,
      "loss": 0.0163,
      "step": 5440
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.6207470297813416,
      "learning_rate": 1.2493522737425176e-05,
      "loss": 0.013,
      "step": 5445
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.2608746290206909,
      "learning_rate": 1.2484588582149559e-05,
      "loss": 0.0094,
      "step": 5450
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.46921461820602417,
      "learning_rate": 1.247565442687394e-05,
      "loss": 0.0125,
      "step": 5455
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.5420220494270325,
      "learning_rate": 1.2466720271598321e-05,
      "loss": 0.0113,
      "step": 5460
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.33676818013191223,
      "learning_rate": 1.2457786116322704e-05,
      "loss": 0.0267,
      "step": 5465
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.7901437878608704,
      "learning_rate": 1.2448851961047083e-05,
      "loss": 0.014,
      "step": 5470
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.610821545124054,
      "learning_rate": 1.2439917805771464e-05,
      "loss": 0.0217,
      "step": 5475
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.3686356544494629,
      "learning_rate": 1.2430983650495847e-05,
      "loss": 0.0217,
      "step": 5480
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.472502738237381,
      "learning_rate": 1.2422049495220228e-05,
      "loss": 0.0106,
      "step": 5485
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.6518288254737854,
      "learning_rate": 1.241311533994461e-05,
      "loss": 0.0127,
      "step": 5490
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.7423241138458252,
      "learning_rate": 1.240418118466899e-05,
      "loss": 0.0152,
      "step": 5495
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.28210926055908203,
      "learning_rate": 1.2395247029393371e-05,
      "loss": 0.0081,
      "step": 5500
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.6291831135749817,
      "learning_rate": 1.2386312874117754e-05,
      "loss": 0.012,
      "step": 5505
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.8124056458473206,
      "learning_rate": 1.2377378718842135e-05,
      "loss": 0.0141,
      "step": 5510
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.3317243158817291,
      "learning_rate": 1.2368444563566516e-05,
      "loss": 0.0067,
      "step": 5515
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.6496483683586121,
      "learning_rate": 1.2359510408290899e-05,
      "loss": 0.0117,
      "step": 5520
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.4485831558704376,
      "learning_rate": 1.2350576253015278e-05,
      "loss": 0.0138,
      "step": 5525
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.6586216688156128,
      "learning_rate": 1.2341642097739659e-05,
      "loss": 0.0245,
      "step": 5530
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5093090534210205,
      "learning_rate": 1.2332707942464042e-05,
      "loss": 0.0127,
      "step": 5535
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5315966606140137,
      "learning_rate": 1.2323773787188423e-05,
      "loss": 0.0129,
      "step": 5540
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.3614465892314911,
      "learning_rate": 1.2314839631912802e-05,
      "loss": 0.0114,
      "step": 5545
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.6309869885444641,
      "learning_rate": 1.2305905476637185e-05,
      "loss": 0.0149,
      "step": 5550
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.6997755169868469,
      "learning_rate": 1.2296971321361566e-05,
      "loss": 0.0107,
      "step": 5555
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5971230864524841,
      "learning_rate": 1.2288037166085947e-05,
      "loss": 0.0108,
      "step": 5560
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.34963735938072205,
      "learning_rate": 1.227910301081033e-05,
      "loss": 0.009,
      "step": 5565
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.48005858063697815,
      "learning_rate": 1.2270168855534709e-05,
      "loss": 0.0097,
      "step": 5570
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.46374374628067017,
      "learning_rate": 1.2261234700259092e-05,
      "loss": 0.0155,
      "step": 5575
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.0261743068695068,
      "learning_rate": 1.2252300544983473e-05,
      "loss": 0.0161,
      "step": 5580
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5249693393707275,
      "learning_rate": 1.2243366389707854e-05,
      "loss": 0.0111,
      "step": 5585
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.6804917454719543,
      "learning_rate": 1.2234432234432237e-05,
      "loss": 0.0137,
      "step": 5590
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.3612353503704071,
      "learning_rate": 1.2225498079156618e-05,
      "loss": 0.0121,
      "step": 5595
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.7740394473075867,
      "learning_rate": 1.2216563923880997e-05,
      "loss": 0.0187,
      "step": 5600
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.6663631796836853,
      "learning_rate": 1.220762976860538e-05,
      "loss": 0.0103,
      "step": 5605
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.4974234402179718,
      "learning_rate": 1.219869561332976e-05,
      "loss": 0.0099,
      "step": 5610
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.5508677959442139,
      "learning_rate": 1.2189761458054142e-05,
      "loss": 0.0208,
      "step": 5615
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.41311079263687134,
      "learning_rate": 1.2180827302778525e-05,
      "loss": 0.0121,
      "step": 5620
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5254323482513428,
      "learning_rate": 1.2171893147502904e-05,
      "loss": 0.0166,
      "step": 5625
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.1438912153244019,
      "learning_rate": 1.2162958992227285e-05,
      "loss": 0.0137,
      "step": 5630
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.25755608081817627,
      "learning_rate": 1.2154024836951668e-05,
      "loss": 0.0294,
      "step": 5635
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.47497978806495667,
      "learning_rate": 1.2145090681676049e-05,
      "loss": 0.0158,
      "step": 5640
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.028806209564209,
      "learning_rate": 1.2136156526400428e-05,
      "loss": 0.0173,
      "step": 5645
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5604512095451355,
      "learning_rate": 1.212722237112481e-05,
      "loss": 0.013,
      "step": 5650
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.015220284461975,
      "learning_rate": 1.2118288215849192e-05,
      "loss": 0.024,
      "step": 5655
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.48487675189971924,
      "learning_rate": 1.2109354060573573e-05,
      "loss": 0.0166,
      "step": 5660
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.28482356667518616,
      "learning_rate": 1.2100419905297956e-05,
      "loss": 0.0126,
      "step": 5665
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.31956416368484497,
      "learning_rate": 1.2091485750022337e-05,
      "loss": 0.0092,
      "step": 5670
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.30106639862060547,
      "learning_rate": 1.208255159474672e-05,
      "loss": 0.0097,
      "step": 5675
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.4242466390132904,
      "learning_rate": 1.2073617439471099e-05,
      "loss": 0.0087,
      "step": 5680
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5314751863479614,
      "learning_rate": 1.206468328419548e-05,
      "loss": 0.0177,
      "step": 5685
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.6850114464759827,
      "learning_rate": 1.2055749128919862e-05,
      "loss": 0.0099,
      "step": 5690
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.4228883385658264,
      "learning_rate": 1.2046814973644244e-05,
      "loss": 0.0105,
      "step": 5695
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.7108424305915833,
      "learning_rate": 1.2037880818368623e-05,
      "loss": 0.0088,
      "step": 5700
    },
    {
      "epoch": 0.46,
      "eval_loss": 0.013003922998905182,
      "eval_pearson_cosine": 0.9397614341232716,
      "eval_pearson_dot": 0.9370161691058996,
      "eval_pearson_euclidean": 0.9290249901323626,
      "eval_pearson_manhattan": 0.9272273324807798,
      "eval_pearson_max": 0.9397614341232716,
      "eval_runtime": 426.205,
      "eval_samples_per_second": 1.173,
      "eval_spearman_cosine": 0.9409476677906711,
      "eval_spearman_dot": 0.9353248292993173,
      "eval_spearman_euclidean": 0.9404788019152075,
      "eval_spearman_manhattan": 0.9399764319057274,
      "eval_spearman_max": 0.9409476677906711,
      "eval_steps_per_second": 1.173,
      "step": 5700
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.4470467269420624,
      "learning_rate": 1.2028946663093006e-05,
      "loss": 0.0196,
      "step": 5705
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.6300716400146484,
      "learning_rate": 1.2020012507817387e-05,
      "loss": 0.0087,
      "step": 5710
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.39396733045578003,
      "learning_rate": 1.2011078352541768e-05,
      "loss": 0.0091,
      "step": 5715
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.46604982018470764,
      "learning_rate": 1.200214419726615e-05,
      "loss": 0.025,
      "step": 5720
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.7758652567863464,
      "learning_rate": 1.1993210041990531e-05,
      "loss": 0.0159,
      "step": 5725
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.7620574235916138,
      "learning_rate": 1.198427588671491e-05,
      "loss": 0.0308,
      "step": 5730
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.474371314048767,
      "learning_rate": 1.1975341731439294e-05,
      "loss": 0.0166,
      "step": 5735
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.13706831634044647,
      "learning_rate": 1.1966407576163675e-05,
      "loss": 0.0129,
      "step": 5740
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.3527317941188812,
      "learning_rate": 1.1957473420888056e-05,
      "loss": 0.0119,
      "step": 5745
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5470644235610962,
      "learning_rate": 1.1948539265612438e-05,
      "loss": 0.0071,
      "step": 5750
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.442442923784256,
      "learning_rate": 1.1939605110336818e-05,
      "loss": 0.0127,
      "step": 5755
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.4580548405647278,
      "learning_rate": 1.19306709550612e-05,
      "loss": 0.0153,
      "step": 5760
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.9100865721702576,
      "learning_rate": 1.1921736799785581e-05,
      "loss": 0.0151,
      "step": 5765
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.5971898436546326,
      "learning_rate": 1.1912802644509962e-05,
      "loss": 0.0091,
      "step": 5770
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.9077457785606384,
      "learning_rate": 1.1903868489234345e-05,
      "loss": 0.015,
      "step": 5775
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.3179706037044525,
      "learning_rate": 1.1894934333958725e-05,
      "loss": 0.0096,
      "step": 5780
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.3894002437591553,
      "learning_rate": 1.1886000178683106e-05,
      "loss": 0.0112,
      "step": 5785
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.24281609058380127,
      "learning_rate": 1.1877066023407488e-05,
      "loss": 0.0082,
      "step": 5790
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.262224018573761,
      "learning_rate": 1.186813186813187e-05,
      "loss": 0.0189,
      "step": 5795
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.38096240162849426,
      "learning_rate": 1.185919771285625e-05,
      "loss": 0.0075,
      "step": 5800
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.299079030752182,
      "learning_rate": 1.1850263557580633e-05,
      "loss": 0.0147,
      "step": 5805
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.43094006180763245,
      "learning_rate": 1.1841329402305012e-05,
      "loss": 0.0121,
      "step": 5810
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.33772242069244385,
      "learning_rate": 1.1832395247029394e-05,
      "loss": 0.0124,
      "step": 5815
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.6729472279548645,
      "learning_rate": 1.1823461091753776e-05,
      "loss": 0.0116,
      "step": 5820
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.858095645904541,
      "learning_rate": 1.1814526936478157e-05,
      "loss": 0.0178,
      "step": 5825
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.4502745270729065,
      "learning_rate": 1.1805592781202537e-05,
      "loss": 0.0123,
      "step": 5830
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.9232989549636841,
      "learning_rate": 1.179665862592692e-05,
      "loss": 0.0194,
      "step": 5835
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.4538329541683197,
      "learning_rate": 1.17877244706513e-05,
      "loss": 0.0205,
      "step": 5840
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.4530210494995117,
      "learning_rate": 1.1778790315375683e-05,
      "loss": 0.0117,
      "step": 5845
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.7247397303581238,
      "learning_rate": 1.1769856160100064e-05,
      "loss": 0.0117,
      "step": 5850
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.6356292963027954,
      "learning_rate": 1.1760922004824444e-05,
      "loss": 0.018,
      "step": 5855
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.48704656958580017,
      "learning_rate": 1.1751987849548826e-05,
      "loss": 0.0128,
      "step": 5860
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5083175301551819,
      "learning_rate": 1.1743053694273207e-05,
      "loss": 0.0127,
      "step": 5865
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.35881999135017395,
      "learning_rate": 1.1734119538997588e-05,
      "loss": 0.0102,
      "step": 5870
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.6129491925239563,
      "learning_rate": 1.1725185383721971e-05,
      "loss": 0.014,
      "step": 5875
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.504448413848877,
      "learning_rate": 1.1716251228446352e-05,
      "loss": 0.0147,
      "step": 5880
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.41546618938446045,
      "learning_rate": 1.1707317073170731e-05,
      "loss": 0.0137,
      "step": 5885
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.3509145677089691,
      "learning_rate": 1.1698382917895114e-05,
      "loss": 0.01,
      "step": 5890
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5603988766670227,
      "learning_rate": 1.1689448762619495e-05,
      "loss": 0.0163,
      "step": 5895
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5339281558990479,
      "learning_rate": 1.1680514607343876e-05,
      "loss": 0.016,
      "step": 5900
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.5996164679527283,
      "learning_rate": 1.1671580452068259e-05,
      "loss": 0.011,
      "step": 5905
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.4414864182472229,
      "learning_rate": 1.1662646296792638e-05,
      "loss": 0.0175,
      "step": 5910
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.43156516551971436,
      "learning_rate": 1.165371214151702e-05,
      "loss": 0.0135,
      "step": 5915
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.6388380527496338,
      "learning_rate": 1.1644777986241402e-05,
      "loss": 0.0149,
      "step": 5920
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.7478085160255432,
      "learning_rate": 1.1635843830965783e-05,
      "loss": 0.0185,
      "step": 5925
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.38189032673835754,
      "learning_rate": 1.1626909675690164e-05,
      "loss": 0.0104,
      "step": 5930
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.4605272114276886,
      "learning_rate": 1.1617975520414547e-05,
      "loss": 0.0093,
      "step": 5935
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.9401006102561951,
      "learning_rate": 1.1609041365138926e-05,
      "loss": 0.01,
      "step": 5940
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.6511146426200867,
      "learning_rate": 1.1600107209863309e-05,
      "loss": 0.0094,
      "step": 5945
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.21353290975093842,
      "learning_rate": 1.159117305458769e-05,
      "loss": 0.0165,
      "step": 5950
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.1418187916278839,
      "learning_rate": 1.1582238899312071e-05,
      "loss": 0.0146,
      "step": 5955
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.268934965133667,
      "learning_rate": 1.1573304744036454e-05,
      "loss": 0.0085,
      "step": 5960
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.8122082948684692,
      "learning_rate": 1.1564370588760833e-05,
      "loss": 0.0155,
      "step": 5965
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.34395936131477356,
      "learning_rate": 1.1555436433485214e-05,
      "loss": 0.013,
      "step": 5970
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.512114405632019,
      "learning_rate": 1.1546502278209597e-05,
      "loss": 0.0116,
      "step": 5975
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.3798539340496063,
      "learning_rate": 1.1537568122933978e-05,
      "loss": 0.0124,
      "step": 5980
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.606410026550293,
      "learning_rate": 1.1528633967658357e-05,
      "loss": 0.0086,
      "step": 5985
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.6842241287231445,
      "learning_rate": 1.151969981238274e-05,
      "loss": 0.0106,
      "step": 5990
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.29534584283828735,
      "learning_rate": 1.1510765657107121e-05,
      "loss": 0.0068,
      "step": 5995
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.3113650381565094,
      "learning_rate": 1.1501831501831502e-05,
      "loss": 0.0155,
      "step": 6000
    },
    {
      "epoch": 0.48,
      "eval_loss": 0.013771675527095795,
      "eval_pearson_cosine": 0.9422620479249941,
      "eval_pearson_dot": 0.9397006206399617,
      "eval_pearson_euclidean": 0.925646983166319,
      "eval_pearson_manhattan": 0.9236079369932951,
      "eval_pearson_max": 0.9422620479249941,
      "eval_runtime": 426.6252,
      "eval_samples_per_second": 1.172,
      "eval_spearman_cosine": 0.9440107520430081,
      "eval_spearman_dot": 0.939121260485042,
      "eval_spearman_euclidean": 0.9447301789207158,
      "eval_spearman_manhattan": 0.9429190356761425,
      "eval_spearman_max": 0.9447301789207158,
      "eval_steps_per_second": 1.172,
      "step": 6000
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5953865647315979,
      "learning_rate": 1.1492897346555885e-05,
      "loss": 0.0088,
      "step": 6005
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.6089617013931274,
      "learning_rate": 1.1483963191280266e-05,
      "loss": 0.0138,
      "step": 6010
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.6975350975990295,
      "learning_rate": 1.1475029036004645e-05,
      "loss": 0.0121,
      "step": 6015
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.6373233199119568,
      "learning_rate": 1.1466094880729028e-05,
      "loss": 0.013,
      "step": 6020
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.4168721139431,
      "learning_rate": 1.1457160725453409e-05,
      "loss": 0.0208,
      "step": 6025
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.9666951298713684,
      "learning_rate": 1.1448226570177792e-05,
      "loss": 0.0151,
      "step": 6030
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.4322957992553711,
      "learning_rate": 1.1439292414902173e-05,
      "loss": 0.0077,
      "step": 6035
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3521668314933777,
      "learning_rate": 1.1430358259626552e-05,
      "loss": 0.0089,
      "step": 6040
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.27815553545951843,
      "learning_rate": 1.1421424104350935e-05,
      "loss": 0.01,
      "step": 6045
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.45623779296875,
      "learning_rate": 1.1412489949075316e-05,
      "loss": 0.0117,
      "step": 6050
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3917030096054077,
      "learning_rate": 1.1403555793799697e-05,
      "loss": 0.0089,
      "step": 6055
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.4462035894393921,
      "learning_rate": 1.139462163852408e-05,
      "loss": 0.0152,
      "step": 6060
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5268591642379761,
      "learning_rate": 1.1385687483248459e-05,
      "loss": 0.0078,
      "step": 6065
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3769688010215759,
      "learning_rate": 1.137675332797284e-05,
      "loss": 0.018,
      "step": 6070
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.003709077835083,
      "learning_rate": 1.1367819172697223e-05,
      "loss": 0.0175,
      "step": 6075
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.336405873298645,
      "learning_rate": 1.1358885017421604e-05,
      "loss": 0.0227,
      "step": 6080
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.4904787540435791,
      "learning_rate": 1.1349950862145985e-05,
      "loss": 0.0117,
      "step": 6085
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.5461546182632446,
      "learning_rate": 1.1341016706870368e-05,
      "loss": 0.0205,
      "step": 6090
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5218220353126526,
      "learning_rate": 1.1332082551594747e-05,
      "loss": 0.0191,
      "step": 6095
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5722084641456604,
      "learning_rate": 1.1323148396319128e-05,
      "loss": 0.015,
      "step": 6100
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.2776796221733093,
      "learning_rate": 1.131421424104351e-05,
      "loss": 0.0115,
      "step": 6105
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.2007933855056763,
      "learning_rate": 1.1305280085767892e-05,
      "loss": 0.017,
      "step": 6110
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.6880375146865845,
      "learning_rate": 1.1296345930492271e-05,
      "loss": 0.0127,
      "step": 6115
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.7174005508422852,
      "learning_rate": 1.1287411775216654e-05,
      "loss": 0.0162,
      "step": 6120
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5776338577270508,
      "learning_rate": 1.1278477619941035e-05,
      "loss": 0.0098,
      "step": 6125
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.0098873376846313,
      "learning_rate": 1.1269543464665418e-05,
      "loss": 0.0161,
      "step": 6130
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.22586317360401154,
      "learning_rate": 1.1260609309389799e-05,
      "loss": 0.0077,
      "step": 6135
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5158787369728088,
      "learning_rate": 1.125167515411418e-05,
      "loss": 0.014,
      "step": 6140
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.8354634046554565,
      "learning_rate": 1.1242740998838562e-05,
      "loss": 0.0132,
      "step": 6145
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.4235319495201111,
      "learning_rate": 1.1233806843562942e-05,
      "loss": 0.0098,
      "step": 6150
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.5751050114631653,
      "learning_rate": 1.1224872688287323e-05,
      "loss": 0.0211,
      "step": 6155
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.681922435760498,
      "learning_rate": 1.1215938533011706e-05,
      "loss": 0.0085,
      "step": 6160
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.48339566588401794,
      "learning_rate": 1.1207004377736087e-05,
      "loss": 0.0161,
      "step": 6165
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5482934713363647,
      "learning_rate": 1.1198070222460466e-05,
      "loss": 0.0087,
      "step": 6170
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.4296235144138336,
      "learning_rate": 1.1189136067184849e-05,
      "loss": 0.0145,
      "step": 6175
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5056319832801819,
      "learning_rate": 1.118020191190923e-05,
      "loss": 0.015,
      "step": 6180
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.7199443578720093,
      "learning_rate": 1.117126775663361e-05,
      "loss": 0.0152,
      "step": 6185
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.2746157646179199,
      "learning_rate": 1.1162333601357993e-05,
      "loss": 0.0108,
      "step": 6190
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.4172118902206421,
      "learning_rate": 1.1153399446082373e-05,
      "loss": 0.0182,
      "step": 6195
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.4746299982070923,
      "learning_rate": 1.1144465290806754e-05,
      "loss": 0.0145,
      "step": 6200
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.581675112247467,
      "learning_rate": 1.1135531135531137e-05,
      "loss": 0.0096,
      "step": 6205
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.39109745621681213,
      "learning_rate": 1.1126596980255518e-05,
      "loss": 0.0143,
      "step": 6210
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.3666527271270752,
      "learning_rate": 1.11176628249799e-05,
      "loss": 0.0104,
      "step": 6215
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.8273961544036865,
      "learning_rate": 1.1108728669704281e-05,
      "loss": 0.0216,
      "step": 6220
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.27460169792175293,
      "learning_rate": 1.109979451442866e-05,
      "loss": 0.0108,
      "step": 6225
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.2216835767030716,
      "learning_rate": 1.1090860359153043e-05,
      "loss": 0.0107,
      "step": 6230
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.36711356043815613,
      "learning_rate": 1.1081926203877425e-05,
      "loss": 0.0114,
      "step": 6235
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.4205690324306488,
      "learning_rate": 1.1072992048601806e-05,
      "loss": 0.0092,
      "step": 6240
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.672114908695221,
      "learning_rate": 1.1064057893326188e-05,
      "loss": 0.0176,
      "step": 6245
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.33661511540412903,
      "learning_rate": 1.1055123738050568e-05,
      "loss": 0.0133,
      "step": 6250
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.6922995448112488,
      "learning_rate": 1.1046189582774949e-05,
      "loss": 0.0161,
      "step": 6255
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5286815166473389,
      "learning_rate": 1.1037255427499331e-05,
      "loss": 0.0161,
      "step": 6260
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.3486752212047577,
      "learning_rate": 1.1028321272223712e-05,
      "loss": 0.0108,
      "step": 6265
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.6005237102508545,
      "learning_rate": 1.1019387116948092e-05,
      "loss": 0.0121,
      "step": 6270
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.47532469034194946,
      "learning_rate": 1.1010452961672475e-05,
      "loss": 0.0292,
      "step": 6275
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.6004465222358704,
      "learning_rate": 1.1001518806396856e-05,
      "loss": 0.0198,
      "step": 6280
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.27914538979530334,
      "learning_rate": 1.0992584651121237e-05,
      "loss": 0.0131,
      "step": 6285
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.4189085066318512,
      "learning_rate": 1.098365049584562e-05,
      "loss": 0.0152,
      "step": 6290
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.46013280749320984,
      "learning_rate": 1.097471634057e-05,
      "loss": 0.0119,
      "step": 6295
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.6204803586006165,
      "learning_rate": 1.096578218529438e-05,
      "loss": 0.0112,
      "step": 6300
    },
    {
      "epoch": 0.51,
      "eval_loss": 0.012779050506651402,
      "eval_pearson_cosine": 0.9423282360276709,
      "eval_pearson_dot": 0.9410928043756498,
      "eval_pearson_euclidean": 0.9267777276223531,
      "eval_pearson_manhattan": 0.9251914910546118,
      "eval_pearson_max": 0.9423282360276709,
      "eval_runtime": 427.1766,
      "eval_samples_per_second": 1.17,
      "eval_spearman_cosine": 0.9424068736274943,
      "eval_spearman_dot": 0.938691754767019,
      "eval_spearman_euclidean": 0.9417786471145886,
      "eval_spearman_manhattan": 0.9395457741830966,
      "eval_spearman_max": 0.9424068736274943,
      "eval_steps_per_second": 1.17,
      "step": 6300
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.3984769284725189,
      "learning_rate": 1.0956848030018762e-05,
      "loss": 0.0113,
      "step": 6305
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.44638150930404663,
      "learning_rate": 1.0947913874743143e-05,
      "loss": 0.0122,
      "step": 6310
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.4902111887931824,
      "learning_rate": 1.0938979719467526e-05,
      "loss": 0.0134,
      "step": 6315
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.32047879695892334,
      "learning_rate": 1.0930045564191907e-05,
      "loss": 0.0147,
      "step": 6320
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.9867672920227051,
      "learning_rate": 1.0921111408916287e-05,
      "loss": 0.0102,
      "step": 6325
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.6817474961280823,
      "learning_rate": 1.091217725364067e-05,
      "loss": 0.0154,
      "step": 6330
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.902030885219574,
      "learning_rate": 1.090324309836505e-05,
      "loss": 0.0111,
      "step": 6335
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.2658748924732208,
      "learning_rate": 1.0894308943089431e-05,
      "loss": 0.0106,
      "step": 6340
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.8799954056739807,
      "learning_rate": 1.0885374787813814e-05,
      "loss": 0.0118,
      "step": 6345
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5085673928260803,
      "learning_rate": 1.0876440632538195e-05,
      "loss": 0.0122,
      "step": 6350
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.2997933626174927,
      "learning_rate": 1.0867506477262575e-05,
      "loss": 0.0217,
      "step": 6355
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5237243175506592,
      "learning_rate": 1.0858572321986957e-05,
      "loss": 0.012,
      "step": 6360
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.2947341799736023,
      "learning_rate": 1.0849638166711338e-05,
      "loss": 0.0064,
      "step": 6365
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.5389357209205627,
      "learning_rate": 1.084070401143572e-05,
      "loss": 0.0105,
      "step": 6370
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.32772096991539,
      "learning_rate": 1.0831769856160102e-05,
      "loss": 0.0102,
      "step": 6375
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.507468044757843,
      "learning_rate": 1.0822835700884481e-05,
      "loss": 0.0081,
      "step": 6380
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.692340612411499,
      "learning_rate": 1.0813901545608862e-05,
      "loss": 0.0247,
      "step": 6385
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.445205956697464,
      "learning_rate": 1.0804967390333245e-05,
      "loss": 0.012,
      "step": 6390
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.7452211976051331,
      "learning_rate": 1.0796033235057626e-05,
      "loss": 0.01,
      "step": 6395
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.355770468711853,
      "learning_rate": 1.0787099079782009e-05,
      "loss": 0.0078,
      "step": 6400
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.6399819254875183,
      "learning_rate": 1.0778164924506388e-05,
      "loss": 0.0114,
      "step": 6405
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.4120681583881378,
      "learning_rate": 1.076923076923077e-05,
      "loss": 0.0061,
      "step": 6410
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.4053487181663513,
      "learning_rate": 1.0760296613955152e-05,
      "loss": 0.0119,
      "step": 6415
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.3720279335975647,
      "learning_rate": 1.0751362458679533e-05,
      "loss": 0.0077,
      "step": 6420
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.6314745545387268,
      "learning_rate": 1.0742428303403914e-05,
      "loss": 0.0089,
      "step": 6425
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.965619683265686,
      "learning_rate": 1.0733494148128297e-05,
      "loss": 0.0152,
      "step": 6430
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5485572218894958,
      "learning_rate": 1.0724559992852676e-05,
      "loss": 0.0104,
      "step": 6435
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.4818495810031891,
      "learning_rate": 1.0715625837577057e-05,
      "loss": 0.0071,
      "step": 6440
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.1216201782226562,
      "learning_rate": 1.070669168230144e-05,
      "loss": 0.0126,
      "step": 6445
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.27217164635658264,
      "learning_rate": 1.0697757527025821e-05,
      "loss": 0.0084,
      "step": 6450
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.4834306240081787,
      "learning_rate": 1.06888233717502e-05,
      "loss": 0.0097,
      "step": 6455
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.5002154111862183,
      "learning_rate": 1.0679889216474583e-05,
      "loss": 0.024,
      "step": 6460
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.35767480731010437,
      "learning_rate": 1.0670955061198964e-05,
      "loss": 0.013,
      "step": 6465
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.43108561635017395,
      "learning_rate": 1.0662020905923345e-05,
      "loss": 0.0122,
      "step": 6470
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.26162779331207275,
      "learning_rate": 1.0653086750647728e-05,
      "loss": 0.0088,
      "step": 6475
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.36457544565200806,
      "learning_rate": 1.0644152595372107e-05,
      "loss": 0.0151,
      "step": 6480
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.32597383856773376,
      "learning_rate": 1.0635218440096492e-05,
      "loss": 0.0103,
      "step": 6485
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.35758069157600403,
      "learning_rate": 1.0626284284820871e-05,
      "loss": 0.008,
      "step": 6490
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.4561416804790497,
      "learning_rate": 1.0617350129545252e-05,
      "loss": 0.0087,
      "step": 6495
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.48591333627700806,
      "learning_rate": 1.0608415974269635e-05,
      "loss": 0.0156,
      "step": 6500
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.42666366696357727,
      "learning_rate": 1.0599481818994016e-05,
      "loss": 0.0153,
      "step": 6505
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.42283836007118225,
      "learning_rate": 1.0590547663718395e-05,
      "loss": 0.0127,
      "step": 6510
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.4533493220806122,
      "learning_rate": 1.0581613508442778e-05,
      "loss": 0.0111,
      "step": 6515
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.49478089809417725,
      "learning_rate": 1.0572679353167159e-05,
      "loss": 0.0143,
      "step": 6520
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5054500102996826,
      "learning_rate": 1.056374519789154e-05,
      "loss": 0.0149,
      "step": 6525
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.460147887468338,
      "learning_rate": 1.0554811042615923e-05,
      "loss": 0.0105,
      "step": 6530
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.31391316652297974,
      "learning_rate": 1.0545876887340302e-05,
      "loss": 0.0075,
      "step": 6535
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.4689933657646179,
      "learning_rate": 1.0536942732064683e-05,
      "loss": 0.015,
      "step": 6540
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.0449711084365845,
      "learning_rate": 1.0528008576789066e-05,
      "loss": 0.007,
      "step": 6545
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.3974328935146332,
      "learning_rate": 1.0519074421513447e-05,
      "loss": 0.014,
      "step": 6550
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.21872344613075256,
      "learning_rate": 1.0510140266237828e-05,
      "loss": 0.0076,
      "step": 6555
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.45412105321884155,
      "learning_rate": 1.050120611096221e-05,
      "loss": 0.0097,
      "step": 6560
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.6600174903869629,
      "learning_rate": 1.049227195568659e-05,
      "loss": 0.0209,
      "step": 6565
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5035395622253418,
      "learning_rate": 1.0483337800410971e-05,
      "loss": 0.0116,
      "step": 6570
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.3042728006839752,
      "learning_rate": 1.0474403645135354e-05,
      "loss": 0.0072,
      "step": 6575
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.8806206583976746,
      "learning_rate": 1.0465469489859735e-05,
      "loss": 0.0138,
      "step": 6580
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.3040008246898651,
      "learning_rate": 1.0456535334584118e-05,
      "loss": 0.0102,
      "step": 6585
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.28891852498054504,
      "learning_rate": 1.0447601179308497e-05,
      "loss": 0.0127,
      "step": 6590
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.4547773003578186,
      "learning_rate": 1.0438667024032878e-05,
      "loss": 0.0186,
      "step": 6595
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.3194609582424164,
      "learning_rate": 1.042973286875726e-05,
      "loss": 0.0091,
      "step": 6600
    },
    {
      "epoch": 0.53,
      "eval_loss": 0.01132145430892706,
      "eval_pearson_cosine": 0.9485659845202755,
      "eval_pearson_dot": 0.9457199499738166,
      "eval_pearson_euclidean": 0.9350354913178518,
      "eval_pearson_manhattan": 0.9332856992885898,
      "eval_pearson_max": 0.9485659845202755,
      "eval_runtime": 426.8248,
      "eval_samples_per_second": 1.171,
      "eval_spearman_cosine": 0.9434670058680236,
      "eval_spearman_dot": 0.9369134436537746,
      "eval_spearman_euclidean": 0.943381469525878,
      "eval_spearman_manhattan": 0.941443701774807,
      "eval_spearman_max": 0.9434670058680236,
      "eval_steps_per_second": 1.171,
      "step": 6600
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.4940083920955658,
      "learning_rate": 1.0420798713481642e-05,
      "loss": 0.0138,
      "step": 6605
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.41744598746299744,
      "learning_rate": 1.0411864558206021e-05,
      "loss": 0.0143,
      "step": 6610
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.9160624146461487,
      "learning_rate": 1.0402930402930404e-05,
      "loss": 0.0263,
      "step": 6615
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.48454248905181885,
      "learning_rate": 1.0393996247654785e-05,
      "loss": 0.0125,
      "step": 6620
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.5490589737892151,
      "learning_rate": 1.0385062092379166e-05,
      "loss": 0.0086,
      "step": 6625
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.471790075302124,
      "learning_rate": 1.0376127937103549e-05,
      "loss": 0.0141,
      "step": 6630
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.862555205821991,
      "learning_rate": 1.036719378182793e-05,
      "loss": 0.0271,
      "step": 6635
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.4942755699157715,
      "learning_rate": 1.0358259626552309e-05,
      "loss": 0.0148,
      "step": 6640
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.6558979749679565,
      "learning_rate": 1.0349325471276692e-05,
      "loss": 0.0169,
      "step": 6645
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.4866179823875427,
      "learning_rate": 1.0340391316001073e-05,
      "loss": 0.0127,
      "step": 6650
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.6332863569259644,
      "learning_rate": 1.0331457160725454e-05,
      "loss": 0.0252,
      "step": 6655
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.2502923309803009,
      "learning_rate": 1.0322523005449837e-05,
      "loss": 0.0071,
      "step": 6660
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.6508748531341553,
      "learning_rate": 1.0313588850174216e-05,
      "loss": 0.0138,
      "step": 6665
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.4118648171424866,
      "learning_rate": 1.0304654694898599e-05,
      "loss": 0.0102,
      "step": 6670
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.3807571828365326,
      "learning_rate": 1.029572053962298e-05,
      "loss": 0.0216,
      "step": 6675
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5610530376434326,
      "learning_rate": 1.028678638434736e-05,
      "loss": 0.0118,
      "step": 6680
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.3441886901855469,
      "learning_rate": 1.0277852229071743e-05,
      "loss": 0.0117,
      "step": 6685
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5432056188583374,
      "learning_rate": 1.0268918073796124e-05,
      "loss": 0.0093,
      "step": 6690
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.4260289967060089,
      "learning_rate": 1.0259983918520504e-05,
      "loss": 0.0126,
      "step": 6695
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.703513503074646,
      "learning_rate": 1.0251049763244887e-05,
      "loss": 0.0146,
      "step": 6700
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5453211069107056,
      "learning_rate": 1.0242115607969268e-05,
      "loss": 0.0102,
      "step": 6705
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.4009830057621002,
      "learning_rate": 1.0233181452693649e-05,
      "loss": 0.01,
      "step": 6710
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.2775367498397827,
      "learning_rate": 1.0224247297418031e-05,
      "loss": 0.0112,
      "step": 6715
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.33914217352867126,
      "learning_rate": 1.021531314214241e-05,
      "loss": 0.0099,
      "step": 6720
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5450855493545532,
      "learning_rate": 1.0206378986866792e-05,
      "loss": 0.0157,
      "step": 6725
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.707729697227478,
      "learning_rate": 1.0197444831591174e-05,
      "loss": 0.012,
      "step": 6730
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.25510841608047485,
      "learning_rate": 1.0188510676315555e-05,
      "loss": 0.0143,
      "step": 6735
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.6719092726707458,
      "learning_rate": 1.0179576521039935e-05,
      "loss": 0.0166,
      "step": 6740
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5788087248802185,
      "learning_rate": 1.0170642365764318e-05,
      "loss": 0.0124,
      "step": 6745
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.6541697382926941,
      "learning_rate": 1.0161708210488699e-05,
      "loss": 0.0105,
      "step": 6750
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.6626589298248291,
      "learning_rate": 1.015277405521308e-05,
      "loss": 0.0076,
      "step": 6755
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.24044017493724823,
      "learning_rate": 1.0143839899937462e-05,
      "loss": 0.0086,
      "step": 6760
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.6562107801437378,
      "learning_rate": 1.0134905744661843e-05,
      "loss": 0.0139,
      "step": 6765
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.48691409826278687,
      "learning_rate": 1.0125971589386226e-05,
      "loss": 0.0095,
      "step": 6770
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.5748230814933777,
      "learning_rate": 1.0117037434110605e-05,
      "loss": 0.0342,
      "step": 6775
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.3513069152832031,
      "learning_rate": 1.0108103278834987e-05,
      "loss": 0.0214,
      "step": 6780
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.2184191197156906,
      "learning_rate": 1.009916912355937e-05,
      "loss": 0.0114,
      "step": 6785
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.3020235300064087,
      "learning_rate": 1.009023496828375e-05,
      "loss": 0.0059,
      "step": 6790
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.3661447763442993,
      "learning_rate": 1.008130081300813e-05,
      "loss": 0.0113,
      "step": 6795
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.29930874705314636,
      "learning_rate": 1.0072366657732512e-05,
      "loss": 0.0095,
      "step": 6800
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.44415831565856934,
      "learning_rate": 1.0063432502456893e-05,
      "loss": 0.0154,
      "step": 6805
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.4747929871082306,
      "learning_rate": 1.0054498347181274e-05,
      "loss": 0.018,
      "step": 6810
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.4862035810947418,
      "learning_rate": 1.0045564191905657e-05,
      "loss": 0.0151,
      "step": 6815
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.8670873641967773,
      "learning_rate": 1.0036630036630037e-05,
      "loss": 0.015,
      "step": 6820
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.4433298408985138,
      "learning_rate": 1.0027695881354418e-05,
      "loss": 0.0091,
      "step": 6825
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.2429455518722534,
      "learning_rate": 1.00187617260788e-05,
      "loss": 0.0107,
      "step": 6830
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.3211395740509033,
      "learning_rate": 1.0009827570803181e-05,
      "loss": 0.0121,
      "step": 6835
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.36794963479042053,
      "learning_rate": 1.0000893415527562e-05,
      "loss": 0.0089,
      "step": 6840
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5613222122192383,
      "learning_rate": 9.991959260251945e-06,
      "loss": 0.0111,
      "step": 6845
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.2746351957321167,
      "learning_rate": 9.983025104976324e-06,
      "loss": 0.0091,
      "step": 6850
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.4677548110485077,
      "learning_rate": 9.974090949700707e-06,
      "loss": 0.0071,
      "step": 6855
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5583066344261169,
      "learning_rate": 9.965156794425088e-06,
      "loss": 0.0108,
      "step": 6860
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.49992015957832336,
      "learning_rate": 9.95622263914947e-06,
      "loss": 0.0113,
      "step": 6865
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.49409008026123047,
      "learning_rate": 9.94728848387385e-06,
      "loss": 0.0087,
      "step": 6870
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.40808090567588806,
      "learning_rate": 9.938354328598231e-06,
      "loss": 0.0153,
      "step": 6875
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.9815873503684998,
      "learning_rate": 9.929420173322614e-06,
      "loss": 0.0075,
      "step": 6880
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.3298436999320984,
      "learning_rate": 9.920486018046993e-06,
      "loss": 0.008,
      "step": 6885
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.3157660961151123,
      "learning_rate": 9.911551862771376e-06,
      "loss": 0.0067,
      "step": 6890
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.6721735000610352,
      "learning_rate": 9.902617707495757e-06,
      "loss": 0.0141,
      "step": 6895
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.2622057795524597,
      "learning_rate": 9.893683552220138e-06,
      "loss": 0.0105,
      "step": 6900
    },
    {
      "epoch": 0.55,
      "eval_loss": 0.011612797155976295,
      "eval_pearson_cosine": 0.9465482224245039,
      "eval_pearson_dot": 0.945316978860193,
      "eval_pearson_euclidean": 0.9291721342489463,
      "eval_pearson_manhattan": 0.9265917577545122,
      "eval_pearson_max": 0.9465482224245039,
      "eval_runtime": 426.2699,
      "eval_samples_per_second": 1.173,
      "eval_spearman_cosine": 0.9449864999459997,
      "eval_spearman_dot": 0.941247380989524,
      "eval_spearman_euclidean": 0.9445647702590809,
      "eval_spearman_manhattan": 0.9420465841863367,
      "eval_spearman_max": 0.9449864999459997,
      "eval_steps_per_second": 1.173,
      "step": 6900
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.6062464714050293,
      "learning_rate": 9.88474939694452e-06,
      "loss": 0.0066,
      "step": 6905
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.7847184538841248,
      "learning_rate": 9.875815241668902e-06,
      "loss": 0.0101,
      "step": 6910
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.28860947489738464,
      "learning_rate": 9.866881086393281e-06,
      "loss": 0.0151,
      "step": 6915
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.6945732831954956,
      "learning_rate": 9.857946931117664e-06,
      "loss": 0.0089,
      "step": 6920
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.6154179573059082,
      "learning_rate": 9.849012775842045e-06,
      "loss": 0.0118,
      "step": 6925
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5512219667434692,
      "learning_rate": 9.840078620566426e-06,
      "loss": 0.0114,
      "step": 6930
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.33222195506095886,
      "learning_rate": 9.831144465290807e-06,
      "loss": 0.0093,
      "step": 6935
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.2524779438972473,
      "learning_rate": 9.822210310015188e-06,
      "loss": 0.014,
      "step": 6940
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.611117959022522,
      "learning_rate": 9.813276154739571e-06,
      "loss": 0.0115,
      "step": 6945
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.9372537732124329,
      "learning_rate": 9.80434199946395e-06,
      "loss": 0.0119,
      "step": 6950
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5313237905502319,
      "learning_rate": 9.795407844188333e-06,
      "loss": 0.0096,
      "step": 6955
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.37667572498321533,
      "learning_rate": 9.786473688912714e-06,
      "loss": 0.0081,
      "step": 6960
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.4171701669692993,
      "learning_rate": 9.777539533637095e-06,
      "loss": 0.0064,
      "step": 6965
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.34626543521881104,
      "learning_rate": 9.768605378361476e-06,
      "loss": 0.0127,
      "step": 6970
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.4644871652126312,
      "learning_rate": 9.759671223085859e-06,
      "loss": 0.0097,
      "step": 6975
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.4637131989002228,
      "learning_rate": 9.75073706781024e-06,
      "loss": 0.0165,
      "step": 6980
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.6172615885734558,
      "learning_rate": 9.741802912534621e-06,
      "loss": 0.0132,
      "step": 6985
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.25134241580963135,
      "learning_rate": 9.732868757259002e-06,
      "loss": 0.0105,
      "step": 6990
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.3779184818267822,
      "learning_rate": 9.723934601983383e-06,
      "loss": 0.0157,
      "step": 6995
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.23347826302051544,
      "learning_rate": 9.715000446707764e-06,
      "loss": 0.0088,
      "step": 7000
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.42765137553215027,
      "learning_rate": 9.706066291432145e-06,
      "loss": 0.0096,
      "step": 7005
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.48463019728660583,
      "learning_rate": 9.697132136156528e-06,
      "loss": 0.0151,
      "step": 7010
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.6167808771133423,
      "learning_rate": 9.688197980880909e-06,
      "loss": 0.008,
      "step": 7015
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.439592570066452,
      "learning_rate": 9.67926382560529e-06,
      "loss": 0.0202,
      "step": 7020
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.3322664797306061,
      "learning_rate": 9.670329670329671e-06,
      "loss": 0.0129,
      "step": 7025
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.4067094922065735,
      "learning_rate": 9.661395515054052e-06,
      "loss": 0.0111,
      "step": 7030
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.7210712432861328,
      "learning_rate": 9.652461359778433e-06,
      "loss": 0.0116,
      "step": 7035
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.9684979319572449,
      "learning_rate": 9.643527204502816e-06,
      "loss": 0.0212,
      "step": 7040
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.6882022619247437,
      "learning_rate": 9.634593049227197e-06,
      "loss": 0.0109,
      "step": 7045
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.42008674144744873,
      "learning_rate": 9.625658893951578e-06,
      "loss": 0.0145,
      "step": 7050
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.3094348907470703,
      "learning_rate": 9.616724738675959e-06,
      "loss": 0.0107,
      "step": 7055
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.18812952935695648,
      "learning_rate": 9.60779058340034e-06,
      "loss": 0.0085,
      "step": 7060
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.6821894645690918,
      "learning_rate": 9.598856428124723e-06,
      "loss": 0.0121,
      "step": 7065
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.9547286033630371,
      "learning_rate": 9.589922272849102e-06,
      "loss": 0.0138,
      "step": 7070
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.8305642604827881,
      "learning_rate": 9.580988117573485e-06,
      "loss": 0.0169,
      "step": 7075
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.28080248832702637,
      "learning_rate": 9.572053962297866e-06,
      "loss": 0.0112,
      "step": 7080
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5253908634185791,
      "learning_rate": 9.563119807022247e-06,
      "loss": 0.0103,
      "step": 7085
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5074714422225952,
      "learning_rate": 9.554185651746628e-06,
      "loss": 0.0102,
      "step": 7090
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.8809208273887634,
      "learning_rate": 9.545251496471009e-06,
      "loss": 0.0164,
      "step": 7095
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.4173926115036011,
      "learning_rate": 9.53631734119539e-06,
      "loss": 0.011,
      "step": 7100
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.39918845891952515,
      "learning_rate": 9.527383185919773e-06,
      "loss": 0.0121,
      "step": 7105
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5598462224006653,
      "learning_rate": 9.518449030644154e-06,
      "loss": 0.0138,
      "step": 7110
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.36775821447372437,
      "learning_rate": 9.509514875368535e-06,
      "loss": 0.0089,
      "step": 7115
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.2542145252227783,
      "learning_rate": 9.500580720092916e-06,
      "loss": 0.0118,
      "step": 7120
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.4689004421234131,
      "learning_rate": 9.491646564817297e-06,
      "loss": 0.0097,
      "step": 7125
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.35405150055885315,
      "learning_rate": 9.48271240954168e-06,
      "loss": 0.0082,
      "step": 7130
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.32674500346183777,
      "learning_rate": 9.473778254266059e-06,
      "loss": 0.0114,
      "step": 7135
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.45037317276000977,
      "learning_rate": 9.464844098990442e-06,
      "loss": 0.0111,
      "step": 7140
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.4101998805999756,
      "learning_rate": 9.455909943714823e-06,
      "loss": 0.0075,
      "step": 7145
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.35887280106544495,
      "learning_rate": 9.446975788439204e-06,
      "loss": 0.0144,
      "step": 7150
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.548945426940918,
      "learning_rate": 9.438041633163585e-06,
      "loss": 0.0106,
      "step": 7155
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.43174099922180176,
      "learning_rate": 9.429107477887966e-06,
      "loss": 0.0114,
      "step": 7160
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5541350841522217,
      "learning_rate": 9.420173322612349e-06,
      "loss": 0.0094,
      "step": 7165
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.3153156638145447,
      "learning_rate": 9.411239167336728e-06,
      "loss": 0.0061,
      "step": 7170
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.7775315046310425,
      "learning_rate": 9.40230501206111e-06,
      "loss": 0.0065,
      "step": 7175
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.34123435616493225,
      "learning_rate": 9.393370856785492e-06,
      "loss": 0.0083,
      "step": 7180
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.6689932942390442,
      "learning_rate": 9.384436701509873e-06,
      "loss": 0.0107,
      "step": 7185
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.4898190200328827,
      "learning_rate": 9.375502546234254e-06,
      "loss": 0.0132,
      "step": 7190
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5248252153396606,
      "learning_rate": 9.366568390958636e-06,
      "loss": 0.0147,
      "step": 7195
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.22730229794979095,
      "learning_rate": 9.357634235683017e-06,
      "loss": 0.0118,
      "step": 7200
    },
    {
      "epoch": 0.58,
      "eval_loss": 0.012877299450337887,
      "eval_pearson_cosine": 0.9490103516617678,
      "eval_pearson_dot": 0.9464473999007177,
      "eval_pearson_euclidean": 0.9406624792498203,
      "eval_pearson_manhattan": 0.9382462132785804,
      "eval_pearson_max": 0.9490103516617678,
      "eval_runtime": 425.5633,
      "eval_samples_per_second": 1.175,
      "eval_spearman_cosine": 0.9464881379525518,
      "eval_spearman_dot": 0.9422772731090924,
      "eval_spearman_euclidean": 0.9460473041892168,
      "eval_spearman_manhattan": 0.9449177636710546,
      "eval_spearman_max": 0.9464881379525518,
      "eval_steps_per_second": 1.175,
      "step": 7200
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.2841651141643524,
      "learning_rate": 9.348700080407399e-06,
      "loss": 0.0056,
      "step": 7205
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.4850158393383026,
      "learning_rate": 9.33976592513178e-06,
      "loss": 0.0191,
      "step": 7210
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5371047854423523,
      "learning_rate": 9.33083176985616e-06,
      "loss": 0.0091,
      "step": 7215
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.4470509886741638,
      "learning_rate": 9.321897614580542e-06,
      "loss": 0.0112,
      "step": 7220
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5325319170951843,
      "learning_rate": 9.312963459304923e-06,
      "loss": 0.0072,
      "step": 7225
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.31553560495376587,
      "learning_rate": 9.304029304029305e-06,
      "loss": 0.0088,
      "step": 7230
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5126965641975403,
      "learning_rate": 9.295095148753685e-06,
      "loss": 0.009,
      "step": 7235
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.6644874215126038,
      "learning_rate": 9.286160993478067e-06,
      "loss": 0.012,
      "step": 7240
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.48291462659835815,
      "learning_rate": 9.277226838202449e-06,
      "loss": 0.011,
      "step": 7245
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.6909160017967224,
      "learning_rate": 9.268292682926831e-06,
      "loss": 0.0178,
      "step": 7250
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.6852349042892456,
      "learning_rate": 9.25935852765121e-06,
      "loss": 0.0153,
      "step": 7255
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5698316097259521,
      "learning_rate": 9.250424372375593e-06,
      "loss": 0.0114,
      "step": 7260
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.15537500381469727,
      "learning_rate": 9.241490217099974e-06,
      "loss": 0.0081,
      "step": 7265
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.26541587710380554,
      "learning_rate": 9.232556061824355e-06,
      "loss": 0.0048,
      "step": 7270
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.491647332906723,
      "learning_rate": 9.223621906548736e-06,
      "loss": 0.0161,
      "step": 7275
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.7400721311569214,
      "learning_rate": 9.214687751273117e-06,
      "loss": 0.0076,
      "step": 7280
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.49569937586784363,
      "learning_rate": 9.2057535959975e-06,
      "loss": 0.0081,
      "step": 7285
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.4648204445838928,
      "learning_rate": 9.19681944072188e-06,
      "loss": 0.0092,
      "step": 7290
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5650725364685059,
      "learning_rate": 9.187885285446262e-06,
      "loss": 0.0072,
      "step": 7295
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.4584939777851105,
      "learning_rate": 9.178951130170643e-06,
      "loss": 0.0122,
      "step": 7300
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.3434280455112457,
      "learning_rate": 9.170016974895024e-06,
      "loss": 0.0081,
      "step": 7305
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.17949648201465607,
      "learning_rate": 9.161082819619405e-06,
      "loss": 0.0076,
      "step": 7310
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.4728049635887146,
      "learning_rate": 9.152148664343788e-06,
      "loss": 0.0166,
      "step": 7315
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5193156599998474,
      "learning_rate": 9.143214509068168e-06,
      "loss": 0.0097,
      "step": 7320
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.8577033877372742,
      "learning_rate": 9.13428035379255e-06,
      "loss": 0.0132,
      "step": 7325
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.3581514060497284,
      "learning_rate": 9.125346198516931e-06,
      "loss": 0.0107,
      "step": 7330
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.3881000876426697,
      "learning_rate": 9.116412043241312e-06,
      "loss": 0.0082,
      "step": 7335
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.0718932151794434,
      "learning_rate": 9.107477887965693e-06,
      "loss": 0.0121,
      "step": 7340
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.44927722215652466,
      "learning_rate": 9.098543732690074e-06,
      "loss": 0.0085,
      "step": 7345
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.47237133979797363,
      "learning_rate": 9.089609577414457e-06,
      "loss": 0.0104,
      "step": 7350
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.4649021625518799,
      "learning_rate": 9.080675422138836e-06,
      "loss": 0.0077,
      "step": 7355
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.365681916475296,
      "learning_rate": 9.07174126686322e-06,
      "loss": 0.0089,
      "step": 7360
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.4933122396469116,
      "learning_rate": 9.0628071115876e-06,
      "loss": 0.0082,
      "step": 7365
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.0981645584106445,
      "learning_rate": 9.053872956311981e-06,
      "loss": 0.0078,
      "step": 7370
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.6281023025512695,
      "learning_rate": 9.044938801036362e-06,
      "loss": 0.0193,
      "step": 7375
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.7304166555404663,
      "learning_rate": 9.036004645760745e-06,
      "loss": 0.0131,
      "step": 7380
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.5784119367599487,
      "learning_rate": 9.027070490485126e-06,
      "loss": 0.0104,
      "step": 7385
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.42180439829826355,
      "learning_rate": 9.018136335209507e-06,
      "loss": 0.0089,
      "step": 7390
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.43457847833633423,
      "learning_rate": 9.009202179933888e-06,
      "loss": 0.0248,
      "step": 7395
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.34278741478919983,
      "learning_rate": 9.00026802465827e-06,
      "loss": 0.0075,
      "step": 7400
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.6066365838050842,
      "learning_rate": 8.99133386938265e-06,
      "loss": 0.0289,
      "step": 7405
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.35957035422325134,
      "learning_rate": 8.982399714107031e-06,
      "loss": 0.0112,
      "step": 7410
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.6378137469291687,
      "learning_rate": 8.973465558831414e-06,
      "loss": 0.0208,
      "step": 7415
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.33062654733657837,
      "learning_rate": 8.964531403555793e-06,
      "loss": 0.0063,
      "step": 7420
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.27983343601226807,
      "learning_rate": 8.955597248280176e-06,
      "loss": 0.0087,
      "step": 7425
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.2651355266571045,
      "learning_rate": 8.946663093004557e-06,
      "loss": 0.0135,
      "step": 7430
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.2877359390258789,
      "learning_rate": 8.937728937728938e-06,
      "loss": 0.0095,
      "step": 7435
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.35000234842300415,
      "learning_rate": 8.92879478245332e-06,
      "loss": 0.0149,
      "step": 7440
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.43265825510025024,
      "learning_rate": 8.9198606271777e-06,
      "loss": 0.0113,
      "step": 7445
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5641202926635742,
      "learning_rate": 8.910926471902083e-06,
      "loss": 0.0108,
      "step": 7450
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.36517974734306335,
      "learning_rate": 8.901992316626464e-06,
      "loss": 0.0142,
      "step": 7455
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.4212518036365509,
      "learning_rate": 8.893058161350845e-06,
      "loss": 0.0081,
      "step": 7460
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.8196327090263367,
      "learning_rate": 8.884124006075226e-06,
      "loss": 0.0167,
      "step": 7465
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5906988978385925,
      "learning_rate": 8.875189850799609e-06,
      "loss": 0.0117,
      "step": 7470
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.4272076189517975,
      "learning_rate": 8.866255695523988e-06,
      "loss": 0.0074,
      "step": 7475
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.2905941903591156,
      "learning_rate": 8.857321540248371e-06,
      "loss": 0.0116,
      "step": 7480
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.409950852394104,
      "learning_rate": 8.848387384972752e-06,
      "loss": 0.0088,
      "step": 7485
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5896927714347839,
      "learning_rate": 8.839453229697133e-06,
      "loss": 0.0102,
      "step": 7490
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.1798476129770279,
      "learning_rate": 8.830519074421514e-06,
      "loss": 0.0092,
      "step": 7495
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.3826355040073395,
      "learning_rate": 8.821584919145895e-06,
      "loss": 0.0144,
      "step": 7500
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.01108083501458168,
      "eval_pearson_cosine": 0.9553236024593702,
      "eval_pearson_dot": 0.9533095646450657,
      "eval_pearson_euclidean": 0.9457066476835266,
      "eval_pearson_manhattan": 0.9435496608448922,
      "eval_pearson_max": 0.9553236024593702,
      "eval_runtime": 425.4174,
      "eval_samples_per_second": 1.175,
      "eval_spearman_cosine": 0.9509822039288156,
      "eval_spearman_dot": 0.947376525506102,
      "eval_spearman_euclidean": 0.9510230040920163,
      "eval_spearman_manhattan": 0.948260017040068,
      "eval_spearman_max": 0.9510230040920163,
      "eval_steps_per_second": 1.175,
      "step": 7500
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.8489444851875305,
      "learning_rate": 8.812650763870276e-06,
      "loss": 0.0235,
      "step": 7505
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.2248634696006775,
      "learning_rate": 8.803716608594657e-06,
      "loss": 0.0052,
      "step": 7510
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.31619253754615784,
      "learning_rate": 8.79478245331904e-06,
      "loss": 0.0088,
      "step": 7515
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.38321641087532043,
      "learning_rate": 8.785848298043421e-06,
      "loss": 0.0097,
      "step": 7520
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3505537211894989,
      "learning_rate": 8.776914142767802e-06,
      "loss": 0.0115,
      "step": 7525
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.20297235250473022,
      "learning_rate": 8.767979987492183e-06,
      "loss": 0.009,
      "step": 7530
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5231132507324219,
      "learning_rate": 8.759045832216566e-06,
      "loss": 0.0125,
      "step": 7535
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.282447874546051,
      "learning_rate": 8.750111676940945e-06,
      "loss": 0.0082,
      "step": 7540
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.1593090146780014,
      "learning_rate": 8.741177521665328e-06,
      "loss": 0.0056,
      "step": 7545
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.19638505578041077,
      "learning_rate": 8.732243366389709e-06,
      "loss": 0.007,
      "step": 7550
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.909408688545227,
      "learning_rate": 8.72330921111409e-06,
      "loss": 0.0111,
      "step": 7555
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.36250588297843933,
      "learning_rate": 8.714375055838471e-06,
      "loss": 0.0087,
      "step": 7560
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.32631489634513855,
      "learning_rate": 8.705440900562852e-06,
      "loss": 0.0096,
      "step": 7565
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.24822083115577698,
      "learning_rate": 8.696506745287235e-06,
      "loss": 0.0089,
      "step": 7570
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.4469005763530731,
      "learning_rate": 8.687572590011614e-06,
      "loss": 0.0108,
      "step": 7575
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5108357071876526,
      "learning_rate": 8.678638434735997e-06,
      "loss": 0.0116,
      "step": 7580
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.4524661600589752,
      "learning_rate": 8.669704279460378e-06,
      "loss": 0.0083,
      "step": 7585
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.627737820148468,
      "learning_rate": 8.660770124184759e-06,
      "loss": 0.0092,
      "step": 7590
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3179764151573181,
      "learning_rate": 8.65183596890914e-06,
      "loss": 0.0087,
      "step": 7595
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.34097155928611755,
      "learning_rate": 8.642901813633523e-06,
      "loss": 0.0071,
      "step": 7600
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.34627482295036316,
      "learning_rate": 8.633967658357904e-06,
      "loss": 0.0103,
      "step": 7605
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.2550590932369232,
      "learning_rate": 8.625033503082285e-06,
      "loss": 0.0125,
      "step": 7610
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3499496877193451,
      "learning_rate": 8.616099347806666e-06,
      "loss": 0.0102,
      "step": 7615
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5095994472503662,
      "learning_rate": 8.607165192531047e-06,
      "loss": 0.0107,
      "step": 7620
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.6900659203529358,
      "learning_rate": 8.598231037255428e-06,
      "loss": 0.0114,
      "step": 7625
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.28145354986190796,
      "learning_rate": 8.589296881979809e-06,
      "loss": 0.01,
      "step": 7630
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3796047866344452,
      "learning_rate": 8.580362726704192e-06,
      "loss": 0.0089,
      "step": 7635
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.3347243368625641,
      "learning_rate": 8.571428571428571e-06,
      "loss": 0.0073,
      "step": 7640
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.34469786286354065,
      "learning_rate": 8.562494416152954e-06,
      "loss": 0.0294,
      "step": 7645
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.6207167506217957,
      "learning_rate": 8.553560260877335e-06,
      "loss": 0.0099,
      "step": 7650
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3273056745529175,
      "learning_rate": 8.544626105601716e-06,
      "loss": 0.0058,
      "step": 7655
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.37536996603012085,
      "learning_rate": 8.535691950326097e-06,
      "loss": 0.0096,
      "step": 7660
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.35086917877197266,
      "learning_rate": 8.52675779505048e-06,
      "loss": 0.0125,
      "step": 7665
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3277010917663574,
      "learning_rate": 8.51782363977486e-06,
      "loss": 0.0089,
      "step": 7670
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.4444209039211273,
      "learning_rate": 8.508889484499242e-06,
      "loss": 0.012,
      "step": 7675
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.22313633561134338,
      "learning_rate": 8.499955329223623e-06,
      "loss": 0.0071,
      "step": 7680
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.38045355677604675,
      "learning_rate": 8.491021173948004e-06,
      "loss": 0.0066,
      "step": 7685
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.25135067105293274,
      "learning_rate": 8.482087018672385e-06,
      "loss": 0.0074,
      "step": 7690
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3080643117427826,
      "learning_rate": 8.473152863396766e-06,
      "loss": 0.0072,
      "step": 7695
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3189115524291992,
      "learning_rate": 8.464218708121148e-06,
      "loss": 0.0122,
      "step": 7700
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.6065992116928101,
      "learning_rate": 8.45528455284553e-06,
      "loss": 0.0099,
      "step": 7705
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.351665198802948,
      "learning_rate": 8.44635039756991e-06,
      "loss": 0.0088,
      "step": 7710
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.2955916225910187,
      "learning_rate": 8.437416242294292e-06,
      "loss": 0.0092,
      "step": 7715
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.43479758501052856,
      "learning_rate": 8.428482087018673e-06,
      "loss": 0.0098,
      "step": 7720
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.7478668689727783,
      "learning_rate": 8.419547931743054e-06,
      "loss": 0.0078,
      "step": 7725
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.2681145966053009,
      "learning_rate": 8.410613776467436e-06,
      "loss": 0.0128,
      "step": 7730
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.472456693649292,
      "learning_rate": 8.401679621191817e-06,
      "loss": 0.0141,
      "step": 7735
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.28826138377189636,
      "learning_rate": 8.392745465916198e-06,
      "loss": 0.0116,
      "step": 7740
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3068752586841583,
      "learning_rate": 8.38381131064058e-06,
      "loss": 0.0084,
      "step": 7745
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.44161757826805115,
      "learning_rate": 8.37487715536496e-06,
      "loss": 0.0092,
      "step": 7750
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.8214365243911743,
      "learning_rate": 8.365943000089343e-06,
      "loss": 0.0097,
      "step": 7755
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.7772095203399658,
      "learning_rate": 8.357008844813723e-06,
      "loss": 0.0096,
      "step": 7760
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.3470393121242523,
      "learning_rate": 8.348074689538105e-06,
      "loss": 0.017,
      "step": 7765
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.6484146118164062,
      "learning_rate": 8.339140534262486e-06,
      "loss": 0.0131,
      "step": 7770
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4481816589832306,
      "learning_rate": 8.330206378986867e-06,
      "loss": 0.0102,
      "step": 7775
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.3032960891723633,
      "learning_rate": 8.321272223711248e-06,
      "loss": 0.0101,
      "step": 7780
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4950280785560608,
      "learning_rate": 8.31233806843563e-06,
      "loss": 0.0118,
      "step": 7785
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.33173587918281555,
      "learning_rate": 8.303403913160012e-06,
      "loss": 0.0077,
      "step": 7790
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.27809616923332214,
      "learning_rate": 8.294469757884393e-06,
      "loss": 0.0081,
      "step": 7795
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.3537757396697998,
      "learning_rate": 8.285535602608774e-06,
      "loss": 0.0086,
      "step": 7800
    },
    {
      "epoch": 0.63,
      "eval_loss": 0.011113057844340801,
      "eval_pearson_cosine": 0.9504736239877235,
      "eval_pearson_dot": 0.94694991283401,
      "eval_pearson_euclidean": 0.9359585895903748,
      "eval_pearson_manhattan": 0.9327258721013392,
      "eval_pearson_max": 0.9504736239877235,
      "eval_runtime": 425.6909,
      "eval_samples_per_second": 1.175,
      "eval_spearman_cosine": 0.9549815799263196,
      "eval_spearman_dot": 0.9496086304345217,
      "eval_spearman_euclidean": 0.955310669242677,
      "eval_spearman_manhattan": 0.9517249588998357,
      "eval_spearman_max": 0.955310669242677,
      "eval_steps_per_second": 1.175,
      "step": 7800
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4825272560119629,
      "learning_rate": 8.276601447333155e-06,
      "loss": 0.0101,
      "step": 7805
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5322403907775879,
      "learning_rate": 8.267667292057536e-06,
      "loss": 0.0151,
      "step": 7810
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.9261417388916016,
      "learning_rate": 8.258733136781917e-06,
      "loss": 0.0159,
      "step": 7815
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.515354573726654,
      "learning_rate": 8.2497989815063e-06,
      "loss": 0.0132,
      "step": 7820
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4399450719356537,
      "learning_rate": 8.24086482623068e-06,
      "loss": 0.0138,
      "step": 7825
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.2733209729194641,
      "learning_rate": 8.231930670955062e-06,
      "loss": 0.009,
      "step": 7830
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.2845936417579651,
      "learning_rate": 8.222996515679443e-06,
      "loss": 0.01,
      "step": 7835
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.27752649784088135,
      "learning_rate": 8.214062360403824e-06,
      "loss": 0.0077,
      "step": 7840
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4586776793003082,
      "learning_rate": 8.205128205128205e-06,
      "loss": 0.0064,
      "step": 7845
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5839340090751648,
      "learning_rate": 8.196194049852586e-06,
      "loss": 0.0107,
      "step": 7850
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4919023811817169,
      "learning_rate": 8.187259894576969e-06,
      "loss": 0.0088,
      "step": 7855
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.3851850926876068,
      "learning_rate": 8.178325739301348e-06,
      "loss": 0.0087,
      "step": 7860
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.46084967255592346,
      "learning_rate": 8.169391584025731e-06,
      "loss": 0.0119,
      "step": 7865
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.792680561542511,
      "learning_rate": 8.160457428750112e-06,
      "loss": 0.0118,
      "step": 7870
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.27852892875671387,
      "learning_rate": 8.151523273474493e-06,
      "loss": 0.0065,
      "step": 7875
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.2725636661052704,
      "learning_rate": 8.142589118198874e-06,
      "loss": 0.007,
      "step": 7880
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4653577208518982,
      "learning_rate": 8.133654962923257e-06,
      "loss": 0.009,
      "step": 7885
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.3033757209777832,
      "learning_rate": 8.124720807647638e-06,
      "loss": 0.0107,
      "step": 7890
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.3090572953224182,
      "learning_rate": 8.115786652372019e-06,
      "loss": 0.0098,
      "step": 7895
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5531060099601746,
      "learning_rate": 8.1068524970964e-06,
      "loss": 0.0126,
      "step": 7900
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.0406345129013062,
      "learning_rate": 8.097918341820781e-06,
      "loss": 0.0189,
      "step": 7905
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.4176005721092224,
      "learning_rate": 8.088984186545162e-06,
      "loss": 0.0076,
      "step": 7910
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5274201035499573,
      "learning_rate": 8.080050031269543e-06,
      "loss": 0.0098,
      "step": 7915
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.29547053575515747,
      "learning_rate": 8.071115875993926e-06,
      "loss": 0.013,
      "step": 7920
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.35848909616470337,
      "learning_rate": 8.062181720718307e-06,
      "loss": 0.0158,
      "step": 7925
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5341098308563232,
      "learning_rate": 8.053247565442688e-06,
      "loss": 0.0076,
      "step": 7930
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.6656579971313477,
      "learning_rate": 8.044313410167069e-06,
      "loss": 0.0104,
      "step": 7935
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5707061290740967,
      "learning_rate": 8.035379254891452e-06,
      "loss": 0.0113,
      "step": 7940
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.4730210304260254,
      "learning_rate": 8.026445099615831e-06,
      "loss": 0.0124,
      "step": 7945
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.2863544225692749,
      "learning_rate": 8.017510944340214e-06,
      "loss": 0.0072,
      "step": 7950
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.8911215662956238,
      "learning_rate": 8.008576789064595e-06,
      "loss": 0.0153,
      "step": 7955
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.34853044152259827,
      "learning_rate": 7.999642633788976e-06,
      "loss": 0.009,
      "step": 7960
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.2679024636745453,
      "learning_rate": 7.990708478513357e-06,
      "loss": 0.0111,
      "step": 7965
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.2996152937412262,
      "learning_rate": 7.981774323237738e-06,
      "loss": 0.0143,
      "step": 7970
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.24828451871871948,
      "learning_rate": 7.97284016796212e-06,
      "loss": 0.0077,
      "step": 7975
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.25271573662757874,
      "learning_rate": 7.9639060126865e-06,
      "loss": 0.0069,
      "step": 7980
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.40006640553474426,
      "learning_rate": 7.954971857410883e-06,
      "loss": 0.0125,
      "step": 7985
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3571016788482666,
      "learning_rate": 7.946037702135264e-06,
      "loss": 0.007,
      "step": 7990
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.2818203270435333,
      "learning_rate": 7.937103546859645e-06,
      "loss": 0.0083,
      "step": 7995
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3765656650066376,
      "learning_rate": 7.928169391584026e-06,
      "loss": 0.0117,
      "step": 8000
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.6141535043716431,
      "learning_rate": 7.919235236308409e-06,
      "loss": 0.009,
      "step": 8005
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.46924063563346863,
      "learning_rate": 7.910301081032788e-06,
      "loss": 0.0142,
      "step": 8010
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5455445051193237,
      "learning_rate": 7.90136692575717e-06,
      "loss": 0.0074,
      "step": 8015
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.49183765053749084,
      "learning_rate": 7.892432770481552e-06,
      "loss": 0.0095,
      "step": 8020
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.21987323462963104,
      "learning_rate": 7.883498615205933e-06,
      "loss": 0.0127,
      "step": 8025
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5709296464920044,
      "learning_rate": 7.874564459930314e-06,
      "loss": 0.0091,
      "step": 8030
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.38452261686325073,
      "learning_rate": 7.865630304654695e-06,
      "loss": 0.0065,
      "step": 8035
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5809503197669983,
      "learning_rate": 7.856696149379078e-06,
      "loss": 0.0159,
      "step": 8040
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.4587538242340088,
      "learning_rate": 7.847761994103457e-06,
      "loss": 0.0089,
      "step": 8045
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.31821274757385254,
      "learning_rate": 7.83882783882784e-06,
      "loss": 0.0104,
      "step": 8050
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.37236472964286804,
      "learning_rate": 7.829893683552221e-06,
      "loss": 0.0132,
      "step": 8055
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.27789589762687683,
      "learning_rate": 7.820959528276602e-06,
      "loss": 0.0089,
      "step": 8060
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.7653221487998962,
      "learning_rate": 7.812025373000983e-06,
      "loss": 0.0082,
      "step": 8065
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5782196521759033,
      "learning_rate": 7.803091217725366e-06,
      "loss": 0.0088,
      "step": 8070
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.2353951632976532,
      "learning_rate": 7.794157062449747e-06,
      "loss": 0.0124,
      "step": 8075
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.3045112192630768,
      "learning_rate": 7.785222907174128e-06,
      "loss": 0.0142,
      "step": 8080
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.1978531926870346,
      "learning_rate": 7.776288751898509e-06,
      "loss": 0.0076,
      "step": 8085
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.31534671783447266,
      "learning_rate": 7.76735459662289e-06,
      "loss": 0.0061,
      "step": 8090
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.38716286420822144,
      "learning_rate": 7.758420441347271e-06,
      "loss": 0.0108,
      "step": 8095
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.3892115354537964,
      "learning_rate": 7.749486286071652e-06,
      "loss": 0.0081,
      "step": 8100
    },
    {
      "epoch": 0.65,
      "eval_loss": 0.009835315868258476,
      "eval_pearson_cosine": 0.9567352891206823,
      "eval_pearson_dot": 0.9527274298242147,
      "eval_pearson_euclidean": 0.9439012020522587,
      "eval_pearson_manhattan": 0.941253071455541,
      "eval_pearson_max": 0.9567352891206823,
      "eval_runtime": 426.1166,
      "eval_samples_per_second": 1.173,
      "eval_spearman_cosine": 0.9577733670934685,
      "eval_spearman_dot": 0.9503731774927098,
      "eval_spearman_euclidean": 0.9585879303517212,
      "eval_spearman_manhattan": 0.9570973323893295,
      "eval_spearman_max": 0.9585879303517212,
      "eval_steps_per_second": 1.173,
      "step": 8100
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.4344653785228729,
      "learning_rate": 7.740552130796035e-06,
      "loss": 0.0153,
      "step": 8105
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.26057007908821106,
      "learning_rate": 7.731617975520416e-06,
      "loss": 0.0061,
      "step": 8110
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.3515275716781616,
      "learning_rate": 7.722683820244797e-06,
      "loss": 0.0118,
      "step": 8115
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.37563014030456543,
      "learning_rate": 7.713749664969178e-06,
      "loss": 0.0088,
      "step": 8120
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.3289440870285034,
      "learning_rate": 7.704815509693559e-06,
      "loss": 0.0112,
      "step": 8125
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.5579915046691895,
      "learning_rate": 7.69588135441794e-06,
      "loss": 0.0182,
      "step": 8130
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.27845335006713867,
      "learning_rate": 7.686947199142321e-06,
      "loss": 0.0069,
      "step": 8135
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.468414843082428,
      "learning_rate": 7.678013043866704e-06,
      "loss": 0.01,
      "step": 8140
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.40395721793174744,
      "learning_rate": 7.669078888591085e-06,
      "loss": 0.0067,
      "step": 8145
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.6070989966392517,
      "learning_rate": 7.660144733315466e-06,
      "loss": 0.0085,
      "step": 8150
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5286884307861328,
      "learning_rate": 7.651210578039847e-06,
      "loss": 0.015,
      "step": 8155
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.17066028714179993,
      "learning_rate": 7.64227642276423e-06,
      "loss": 0.0094,
      "step": 8160
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.3274366855621338,
      "learning_rate": 7.633342267488609e-06,
      "loss": 0.0102,
      "step": 8165
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.4702674448490143,
      "learning_rate": 7.624408112212991e-06,
      "loss": 0.0081,
      "step": 8170
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5972501635551453,
      "learning_rate": 7.6154739569373726e-06,
      "loss": 0.0083,
      "step": 8175
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.38488277792930603,
      "learning_rate": 7.606539801661753e-06,
      "loss": 0.0149,
      "step": 8180
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.4285413920879364,
      "learning_rate": 7.597605646386135e-06,
      "loss": 0.0163,
      "step": 8185
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.43292173743247986,
      "learning_rate": 7.5886714911105165e-06,
      "loss": 0.009,
      "step": 8190
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.7645611763000488,
      "learning_rate": 7.579737335834897e-06,
      "loss": 0.0131,
      "step": 8195
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.3122585415840149,
      "learning_rate": 7.570803180559279e-06,
      "loss": 0.0206,
      "step": 8200
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.6754368543624878,
      "learning_rate": 7.5618690252836605e-06,
      "loss": 0.0078,
      "step": 8205
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.8362284302711487,
      "learning_rate": 7.5529348700080415e-06,
      "loss": 0.0114,
      "step": 8210
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.3399483263492584,
      "learning_rate": 7.5440007147324226e-06,
      "loss": 0.0102,
      "step": 8215
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.3838357627391815,
      "learning_rate": 7.535066559456804e-06,
      "loss": 0.0106,
      "step": 8220
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.390297532081604,
      "learning_rate": 7.5261324041811855e-06,
      "loss": 0.0124,
      "step": 8225
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.45605188608169556,
      "learning_rate": 7.5171982489055665e-06,
      "loss": 0.0067,
      "step": 8230
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.6078754663467407,
      "learning_rate": 7.5082640936299476e-06,
      "loss": 0.0089,
      "step": 8235
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.4810434877872467,
      "learning_rate": 7.4993299383543294e-06,
      "loss": 0.0155,
      "step": 8240
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.4500656723976135,
      "learning_rate": 7.490395783078711e-06,
      "loss": 0.0119,
      "step": 8245
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.35004329681396484,
      "learning_rate": 7.4814616278030915e-06,
      "loss": 0.0116,
      "step": 8250
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.45475873351097107,
      "learning_rate": 7.472527472527473e-06,
      "loss": 0.013,
      "step": 8255
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.4054245352745056,
      "learning_rate": 7.463593317251855e-06,
      "loss": 0.0095,
      "step": 8260
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.23859477043151855,
      "learning_rate": 7.4546591619762355e-06,
      "loss": 0.0053,
      "step": 8265
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.2994462251663208,
      "learning_rate": 7.445725006700617e-06,
      "loss": 0.0061,
      "step": 8270
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.34221136569976807,
      "learning_rate": 7.436790851424998e-06,
      "loss": 0.0084,
      "step": 8275
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.7677508592605591,
      "learning_rate": 7.4278566961493794e-06,
      "loss": 0.0076,
      "step": 8280
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.3268228769302368,
      "learning_rate": 7.4189225408737605e-06,
      "loss": 0.0082,
      "step": 8285
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.36191943287849426,
      "learning_rate": 7.409988385598142e-06,
      "loss": 0.0071,
      "step": 8290
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.7258814573287964,
      "learning_rate": 7.401054230322524e-06,
      "loss": 0.0182,
      "step": 8295
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.2820124626159668,
      "learning_rate": 7.3921200750469045e-06,
      "loss": 0.0088,
      "step": 8300
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.459287166595459,
      "learning_rate": 7.383185919771286e-06,
      "loss": 0.0114,
      "step": 8305
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.33336150646209717,
      "learning_rate": 7.374251764495668e-06,
      "loss": 0.0101,
      "step": 8310
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.47036316990852356,
      "learning_rate": 7.365317609220048e-06,
      "loss": 0.0105,
      "step": 8315
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.29560941457748413,
      "learning_rate": 7.35638345394443e-06,
      "loss": 0.008,
      "step": 8320
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.6284127235412598,
      "learning_rate": 7.347449298668811e-06,
      "loss": 0.01,
      "step": 8325
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.46273162961006165,
      "learning_rate": 7.338515143393192e-06,
      "loss": 0.0092,
      "step": 8330
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.4691775143146515,
      "learning_rate": 7.329580988117574e-06,
      "loss": 0.0095,
      "step": 8335
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.6613128185272217,
      "learning_rate": 7.320646832841955e-06,
      "loss": 0.0147,
      "step": 8340
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.6718743443489075,
      "learning_rate": 7.311712677566337e-06,
      "loss": 0.0138,
      "step": 8345
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.33338358998298645,
      "learning_rate": 7.302778522290717e-06,
      "loss": 0.0079,
      "step": 8350
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.1531884670257568,
      "learning_rate": 7.293844367015099e-06,
      "loss": 0.0115,
      "step": 8355
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.598302960395813,
      "learning_rate": 7.284910211739481e-06,
      "loss": 0.0165,
      "step": 8360
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.385075181722641,
      "learning_rate": 7.275976056463861e-06,
      "loss": 0.0098,
      "step": 8365
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5176331400871277,
      "learning_rate": 7.267041901188243e-06,
      "loss": 0.0096,
      "step": 8370
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.1693544238805771,
      "learning_rate": 7.258107745912625e-06,
      "loss": 0.0144,
      "step": 8375
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.514583170413971,
      "learning_rate": 7.249173590637005e-06,
      "loss": 0.0138,
      "step": 8380
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.3841685950756073,
      "learning_rate": 7.240239435361387e-06,
      "loss": 0.0087,
      "step": 8385
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.29203617572784424,
      "learning_rate": 7.231305280085768e-06,
      "loss": 0.0065,
      "step": 8390
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.3199155926704407,
      "learning_rate": 7.22237112481015e-06,
      "loss": 0.0091,
      "step": 8395
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.42012354731559753,
      "learning_rate": 7.213436969534531e-06,
      "loss": 0.0099,
      "step": 8400
    },
    {
      "epoch": 0.68,
      "eval_loss": 0.010101383551955223,
      "eval_pearson_cosine": 0.9563888367109727,
      "eval_pearson_dot": 0.9545547272029818,
      "eval_pearson_euclidean": 0.9477205365409502,
      "eval_pearson_manhattan": 0.9455823227770966,
      "eval_pearson_max": 0.9563888367109727,
      "eval_runtime": 425.8741,
      "eval_samples_per_second": 1.174,
      "eval_spearman_cosine": 0.957809943239773,
      "eval_spearman_dot": 0.9524725138900555,
      "eval_spearman_euclidean": 0.9574677018708074,
      "eval_spearman_manhattan": 0.955451117804471,
      "eval_spearman_max": 0.957809943239773,
      "eval_steps_per_second": 1.174,
      "step": 8400
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.652422308921814,
      "learning_rate": 7.204502814258912e-06,
      "loss": 0.0135,
      "step": 8405
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.32259654998779297,
      "learning_rate": 7.195568658983294e-06,
      "loss": 0.0065,
      "step": 8410
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.4164605140686035,
      "learning_rate": 7.186634503707674e-06,
      "loss": 0.0094,
      "step": 8415
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.27093705534935,
      "learning_rate": 7.177700348432056e-06,
      "loss": 0.0076,
      "step": 8420
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.3182103633880615,
      "learning_rate": 7.168766193156438e-06,
      "loss": 0.0094,
      "step": 8425
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.38074642419815063,
      "learning_rate": 7.159832037880819e-06,
      "loss": 0.0088,
      "step": 8430
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.3600986897945404,
      "learning_rate": 7.1508978826052e-06,
      "loss": 0.0085,
      "step": 8435
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.6063751578330994,
      "learning_rate": 7.141963727329582e-06,
      "loss": 0.0109,
      "step": 8440
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.41850802302360535,
      "learning_rate": 7.133029572053963e-06,
      "loss": 0.0079,
      "step": 8445
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.27119773626327515,
      "learning_rate": 7.124095416778344e-06,
      "loss": 0.0063,
      "step": 8450
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.43539103865623474,
      "learning_rate": 7.115161261502725e-06,
      "loss": 0.0074,
      "step": 8455
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.20246903598308563,
      "learning_rate": 7.106227106227107e-06,
      "loss": 0.0078,
      "step": 8460
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.628633975982666,
      "learning_rate": 7.097292950951488e-06,
      "loss": 0.0103,
      "step": 8465
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.22282055020332336,
      "learning_rate": 7.088358795675869e-06,
      "loss": 0.007,
      "step": 8470
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.37223970890045166,
      "learning_rate": 7.079424640400251e-06,
      "loss": 0.0082,
      "step": 8475
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.36613044142723083,
      "learning_rate": 7.070490485124633e-06,
      "loss": 0.0102,
      "step": 8480
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5507646203041077,
      "learning_rate": 7.061556329849013e-06,
      "loss": 0.01,
      "step": 8485
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.47819945216178894,
      "learning_rate": 7.052622174573395e-06,
      "loss": 0.0074,
      "step": 8490
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.6937453150749207,
      "learning_rate": 7.043688019297776e-06,
      "loss": 0.0072,
      "step": 8495
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.3023397624492645,
      "learning_rate": 7.034753864022157e-06,
      "loss": 0.0093,
      "step": 8500
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.35632213950157166,
      "learning_rate": 7.025819708746539e-06,
      "loss": 0.012,
      "step": 8505
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.3277115821838379,
      "learning_rate": 7.01688555347092e-06,
      "loss": 0.0077,
      "step": 8510
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5197334289550781,
      "learning_rate": 7.007951398195301e-06,
      "loss": 0.0063,
      "step": 8515
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.23288393020629883,
      "learning_rate": 6.999017242919682e-06,
      "loss": 0.0066,
      "step": 8520
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.36602354049682617,
      "learning_rate": 6.990083087644064e-06,
      "loss": 0.0084,
      "step": 8525
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.34479185938835144,
      "learning_rate": 6.981148932368446e-06,
      "loss": 0.0105,
      "step": 8530
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.37236830592155457,
      "learning_rate": 6.972214777092826e-06,
      "loss": 0.0092,
      "step": 8535
    },
    {
      "epoch": 0.69,
      "grad_norm": 1.0044472217559814,
      "learning_rate": 6.963280621817208e-06,
      "loss": 0.0157,
      "step": 8540
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5638419389724731,
      "learning_rate": 6.95434646654159e-06,
      "loss": 0.0092,
      "step": 8545
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.991611123085022,
      "learning_rate": 6.94541231126597e-06,
      "loss": 0.0151,
      "step": 8550
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.6007388830184937,
      "learning_rate": 6.936478155990352e-06,
      "loss": 0.0111,
      "step": 8555
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5254615545272827,
      "learning_rate": 6.927544000714733e-06,
      "loss": 0.0095,
      "step": 8560
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.429267019033432,
      "learning_rate": 6.918609845439115e-06,
      "loss": 0.0093,
      "step": 8565
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5293323397636414,
      "learning_rate": 6.909675690163496e-06,
      "loss": 0.0139,
      "step": 8570
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.375843346118927,
      "learning_rate": 6.900741534887877e-06,
      "loss": 0.0052,
      "step": 8575
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.36588814854621887,
      "learning_rate": 6.891807379612259e-06,
      "loss": 0.0087,
      "step": 8580
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.2552180290222168,
      "learning_rate": 6.882873224336639e-06,
      "loss": 0.0073,
      "step": 8585
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.4466809332370758,
      "learning_rate": 6.873939069061021e-06,
      "loss": 0.0138,
      "step": 8590
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.36056727170944214,
      "learning_rate": 6.865004913785403e-06,
      "loss": 0.0074,
      "step": 8595
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.542684018611908,
      "learning_rate": 6.856070758509783e-06,
      "loss": 0.011,
      "step": 8600
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.6635081171989441,
      "learning_rate": 6.847136603234165e-06,
      "loss": 0.0153,
      "step": 8605
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.453122615814209,
      "learning_rate": 6.838202447958547e-06,
      "loss": 0.0078,
      "step": 8610
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.4757082760334015,
      "learning_rate": 6.829268292682928e-06,
      "loss": 0.0074,
      "step": 8615
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.4712415337562561,
      "learning_rate": 6.820334137407309e-06,
      "loss": 0.0084,
      "step": 8620
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5353173613548279,
      "learning_rate": 6.81139998213169e-06,
      "loss": 0.0116,
      "step": 8625
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.14257650077342987,
      "learning_rate": 6.802465826856072e-06,
      "loss": 0.0056,
      "step": 8630
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5088415145874023,
      "learning_rate": 6.793531671580452e-06,
      "loss": 0.0112,
      "step": 8635
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.6079709529876709,
      "learning_rate": 6.784597516304834e-06,
      "loss": 0.0071,
      "step": 8640
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.32311755418777466,
      "learning_rate": 6.775663361029216e-06,
      "loss": 0.0228,
      "step": 8645
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.7822017073631287,
      "learning_rate": 6.766729205753596e-06,
      "loss": 0.0139,
      "step": 8650
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.3916921615600586,
      "learning_rate": 6.757795050477978e-06,
      "loss": 0.0077,
      "step": 8655
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5070155262947083,
      "learning_rate": 6.7488608952023596e-06,
      "loss": 0.0088,
      "step": 8660
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.38798630237579346,
      "learning_rate": 6.739926739926741e-06,
      "loss": 0.0072,
      "step": 8665
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.518268883228302,
      "learning_rate": 6.730992584651122e-06,
      "loss": 0.0056,
      "step": 8670
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.28463345766067505,
      "learning_rate": 6.7220584293755035e-06,
      "loss": 0.0072,
      "step": 8675
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.365010529756546,
      "learning_rate": 6.7131242740998846e-06,
      "loss": 0.0073,
      "step": 8680
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.17222385108470917,
      "learning_rate": 6.704190118824266e-06,
      "loss": 0.0125,
      "step": 8685
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.45439592003822327,
      "learning_rate": 6.695255963548647e-06,
      "loss": 0.0101,
      "step": 8690
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.35715913772583,
      "learning_rate": 6.6863218082730285e-06,
      "loss": 0.0081,
      "step": 8695
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.27602097392082214,
      "learning_rate": 6.677387652997409e-06,
      "loss": 0.0108,
      "step": 8700
    },
    {
      "epoch": 0.7,
      "eval_loss": 0.009161743335425854,
      "eval_pearson_cosine": 0.9580263482961653,
      "eval_pearson_dot": 0.9559365737783729,
      "eval_pearson_euclidean": 0.9478635761936823,
      "eval_pearson_manhattan": 0.9463702679634486,
      "eval_pearson_max": 0.9580263482961653,
      "eval_runtime": 425.8531,
      "eval_samples_per_second": 1.174,
      "eval_spearman_cosine": 0.9578528554114217,
      "eval_spearman_dot": 0.9526426265705062,
      "eval_spearman_euclidean": 0.9572422929691717,
      "eval_spearman_manhattan": 0.9564781299125197,
      "eval_spearman_max": 0.9578528554114217,
      "eval_steps_per_second": 1.174,
      "step": 8700
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.22529442608356476,
      "learning_rate": 6.668453497721791e-06,
      "loss": 0.0061,
      "step": 8705
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.21936196088790894,
      "learning_rate": 6.6595193424461725e-06,
      "loss": 0.0113,
      "step": 8710
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.36036255955696106,
      "learning_rate": 6.650585187170554e-06,
      "loss": 0.0108,
      "step": 8715
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.4151296615600586,
      "learning_rate": 6.6416510318949346e-06,
      "loss": 0.0067,
      "step": 8720
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.44289371371269226,
      "learning_rate": 6.6327168766193165e-06,
      "loss": 0.0078,
      "step": 8725
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.5130812525749207,
      "learning_rate": 6.6237827213436975e-06,
      "loss": 0.0093,
      "step": 8730
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.32452791929244995,
      "learning_rate": 6.6148485660680785e-06,
      "loss": 0.0121,
      "step": 8735
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.2943861484527588,
      "learning_rate": 6.6059144107924596e-06,
      "loss": 0.0062,
      "step": 8740
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.6451315879821777,
      "learning_rate": 6.5969802555168415e-06,
      "loss": 0.0078,
      "step": 8745
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.496124804019928,
      "learning_rate": 6.588046100241223e-06,
      "loss": 0.0088,
      "step": 8750
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.4250244200229645,
      "learning_rate": 6.5791119449656035e-06,
      "loss": 0.008,
      "step": 8755
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.35763731598854065,
      "learning_rate": 6.570177789689985e-06,
      "loss": 0.0139,
      "step": 8760
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.3363899290561676,
      "learning_rate": 6.561243634414367e-06,
      "loss": 0.0074,
      "step": 8765
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.21450985968112946,
      "learning_rate": 6.5523094791387475e-06,
      "loss": 0.0072,
      "step": 8770
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.62349534034729,
      "learning_rate": 6.543375323863129e-06,
      "loss": 0.0089,
      "step": 8775
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.3837112784385681,
      "learning_rate": 6.534441168587511e-06,
      "loss": 0.0088,
      "step": 8780
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.3757342994213104,
      "learning_rate": 6.5255070133118915e-06,
      "loss": 0.0127,
      "step": 8785
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.6816122531890869,
      "learning_rate": 6.516572858036273e-06,
      "loss": 0.0128,
      "step": 8790
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.46092236042022705,
      "learning_rate": 6.507638702760654e-06,
      "loss": 0.0063,
      "step": 8795
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.24193042516708374,
      "learning_rate": 6.498704547485036e-06,
      "loss": 0.0089,
      "step": 8800
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.41303420066833496,
      "learning_rate": 6.4897703922094165e-06,
      "loss": 0.0084,
      "step": 8805
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.4693982005119324,
      "learning_rate": 6.480836236933798e-06,
      "loss": 0.0113,
      "step": 8810
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.37234067916870117,
      "learning_rate": 6.47190208165818e-06,
      "loss": 0.0085,
      "step": 8815
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.27188965678215027,
      "learning_rate": 6.4629679263825604e-06,
      "loss": 0.0088,
      "step": 8820
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.42703506350517273,
      "learning_rate": 6.454033771106942e-06,
      "loss": 0.0051,
      "step": 8825
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.42156967520713806,
      "learning_rate": 6.445099615831324e-06,
      "loss": 0.0056,
      "step": 8830
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.6847637891769409,
      "learning_rate": 6.436165460555704e-06,
      "loss": 0.0257,
      "step": 8835
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5083047151565552,
      "learning_rate": 6.427231305280086e-06,
      "loss": 0.0142,
      "step": 8840
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5095330476760864,
      "learning_rate": 6.418297150004467e-06,
      "loss": 0.0086,
      "step": 8845
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.42313599586486816,
      "learning_rate": 6.409362994728849e-06,
      "loss": 0.0114,
      "step": 8850
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.49969157576560974,
      "learning_rate": 6.40042883945323e-06,
      "loss": 0.0072,
      "step": 8855
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5221359133720398,
      "learning_rate": 6.391494684177611e-06,
      "loss": 0.0087,
      "step": 8860
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5581823587417603,
      "learning_rate": 6.382560528901993e-06,
      "loss": 0.0236,
      "step": 8865
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.4249207079410553,
      "learning_rate": 6.373626373626373e-06,
      "loss": 0.0109,
      "step": 8870
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.4535931646823883,
      "learning_rate": 6.364692218350755e-06,
      "loss": 0.01,
      "step": 8875
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.2545453906059265,
      "learning_rate": 6.355758063075137e-06,
      "loss": 0.0065,
      "step": 8880
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.22866855561733246,
      "learning_rate": 6.346823907799519e-06,
      "loss": 0.0093,
      "step": 8885
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.3718758523464203,
      "learning_rate": 6.337889752523899e-06,
      "loss": 0.0112,
      "step": 8890
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.41113823652267456,
      "learning_rate": 6.328955597248281e-06,
      "loss": 0.0087,
      "step": 8895
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.3951786160469055,
      "learning_rate": 6.320021441972662e-06,
      "loss": 0.0053,
      "step": 8900
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.2985374927520752,
      "learning_rate": 6.311087286697043e-06,
      "loss": 0.0065,
      "step": 8905
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.4587487578392029,
      "learning_rate": 6.302153131421424e-06,
      "loss": 0.0107,
      "step": 8910
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.3964894115924835,
      "learning_rate": 6.293218976145806e-06,
      "loss": 0.0091,
      "step": 8915
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.3453603684902191,
      "learning_rate": 6.284284820870187e-06,
      "loss": 0.0138,
      "step": 8920
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.3030305802822113,
      "learning_rate": 6.275350665594568e-06,
      "loss": 0.0073,
      "step": 8925
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.4225490391254425,
      "learning_rate": 6.26641651031895e-06,
      "loss": 0.0114,
      "step": 8930
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.48078760504722595,
      "learning_rate": 6.257482355043332e-06,
      "loss": 0.0084,
      "step": 8935
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.33763763308525085,
      "learning_rate": 6.248548199767712e-06,
      "loss": 0.0075,
      "step": 8940
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5060855746269226,
      "learning_rate": 6.239614044492094e-06,
      "loss": 0.0163,
      "step": 8945
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.4293050765991211,
      "learning_rate": 6.230679889216476e-06,
      "loss": 0.0108,
      "step": 8950
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5132399201393127,
      "learning_rate": 6.221745733940856e-06,
      "loss": 0.0067,
      "step": 8955
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.3775080144405365,
      "learning_rate": 6.212811578665238e-06,
      "loss": 0.0096,
      "step": 8960
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.4786341190338135,
      "learning_rate": 6.203877423389619e-06,
      "loss": 0.0158,
      "step": 8965
    },
    {
      "epoch": 0.72,
      "grad_norm": 2.233818292617798,
      "learning_rate": 6.194943268114e-06,
      "loss": 0.0312,
      "step": 8970
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.41451793909072876,
      "learning_rate": 6.186009112838381e-06,
      "loss": 0.0062,
      "step": 8975
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.7258878946304321,
      "learning_rate": 6.177074957562763e-06,
      "loss": 0.0157,
      "step": 8980
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.18429698050022125,
      "learning_rate": 6.168140802287145e-06,
      "loss": 0.0079,
      "step": 8985
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.457084983587265,
      "learning_rate": 6.159206647011525e-06,
      "loss": 0.0065,
      "step": 8990
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.315451979637146,
      "learning_rate": 6.150272491735907e-06,
      "loss": 0.0116,
      "step": 8995
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.7770200371742249,
      "learning_rate": 6.141338336460289e-06,
      "loss": 0.0117,
      "step": 9000
    },
    {
      "epoch": 0.72,
      "eval_loss": 0.008854083716869354,
      "eval_pearson_cosine": 0.9601944529575254,
      "eval_pearson_dot": 0.9579298722455568,
      "eval_pearson_euclidean": 0.949392137595604,
      "eval_pearson_manhattan": 0.9478896448618476,
      "eval_pearson_max": 0.9601944529575254,
      "eval_runtime": 426.0015,
      "eval_samples_per_second": 1.174,
      "eval_spearman_cosine": 0.9579289837159349,
      "eval_spearman_dot": 0.9529132516530067,
      "eval_spearman_euclidean": 0.9576313825255303,
      "eval_spearman_manhattan": 0.9561580646322584,
      "eval_spearman_max": 0.9579289837159349,
      "eval_steps_per_second": 1.174,
      "step": 9000
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5715427994728088,
      "learning_rate": 6.132404181184669e-06,
      "loss": 0.0099,
      "step": 9005
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.9296368956565857,
      "learning_rate": 6.123470025909051e-06,
      "loss": 0.0204,
      "step": 9010
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.4095558226108551,
      "learning_rate": 6.114535870633432e-06,
      "loss": 0.0089,
      "step": 9015
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5069900751113892,
      "learning_rate": 6.105601715357813e-06,
      "loss": 0.0089,
      "step": 9020
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.44209763407707214,
      "learning_rate": 6.096667560082195e-06,
      "loss": 0.0096,
      "step": 9025
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.3158562183380127,
      "learning_rate": 6.087733404806576e-06,
      "loss": 0.0106,
      "step": 9030
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.516979992389679,
      "learning_rate": 6.078799249530958e-06,
      "loss": 0.0138,
      "step": 9035
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.1376364678144455,
      "learning_rate": 6.069865094255338e-06,
      "loss": 0.0073,
      "step": 9040
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.38684171438217163,
      "learning_rate": 6.06093093897972e-06,
      "loss": 0.0084,
      "step": 9045
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.2780005633831024,
      "learning_rate": 6.051996783704102e-06,
      "loss": 0.0095,
      "step": 9050
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.4031628668308258,
      "learning_rate": 6.043062628428482e-06,
      "loss": 0.0096,
      "step": 9055
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.24591538310050964,
      "learning_rate": 6.034128473152864e-06,
      "loss": 0.006,
      "step": 9060
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.42805585265159607,
      "learning_rate": 6.025194317877246e-06,
      "loss": 0.0073,
      "step": 9065
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.37655743956565857,
      "learning_rate": 6.016260162601627e-06,
      "loss": 0.0058,
      "step": 9070
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.4816523790359497,
      "learning_rate": 6.007326007326008e-06,
      "loss": 0.0116,
      "step": 9075
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.5275557637214661,
      "learning_rate": 5.998391852050389e-06,
      "loss": 0.0108,
      "step": 9080
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.6304175853729248,
      "learning_rate": 5.989457696774771e-06,
      "loss": 0.0084,
      "step": 9085
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.24132008850574493,
      "learning_rate": 5.980523541499152e-06,
      "loss": 0.0061,
      "step": 9090
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.22209247946739197,
      "learning_rate": 5.971589386223533e-06,
      "loss": 0.0058,
      "step": 9095
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.7451222538948059,
      "learning_rate": 5.962655230947915e-06,
      "loss": 0.0084,
      "step": 9100
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.31591641902923584,
      "learning_rate": 5.953721075672295e-06,
      "loss": 0.0078,
      "step": 9105
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.2269444316625595,
      "learning_rate": 5.944786920396677e-06,
      "loss": 0.0053,
      "step": 9110
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.3011891841888428,
      "learning_rate": 5.935852765121059e-06,
      "loss": 0.0078,
      "step": 9115
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.4496324360370636,
      "learning_rate": 5.92691860984544e-06,
      "loss": 0.0069,
      "step": 9120
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.3193652927875519,
      "learning_rate": 5.917984454569821e-06,
      "loss": 0.0096,
      "step": 9125
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.20983242988586426,
      "learning_rate": 5.909050299294203e-06,
      "loss": 0.0062,
      "step": 9130
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.32339081168174744,
      "learning_rate": 5.900116144018584e-06,
      "loss": 0.0103,
      "step": 9135
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.4473322927951813,
      "learning_rate": 5.891181988742965e-06,
      "loss": 0.0069,
      "step": 9140
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.3048810660839081,
      "learning_rate": 5.882247833467346e-06,
      "loss": 0.0087,
      "step": 9145
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5159117579460144,
      "learning_rate": 5.873313678191728e-06,
      "loss": 0.007,
      "step": 9150
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.7469659447669983,
      "learning_rate": 5.864379522916109e-06,
      "loss": 0.0127,
      "step": 9155
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.36824101209640503,
      "learning_rate": 5.85544536764049e-06,
      "loss": 0.0083,
      "step": 9160
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5489497184753418,
      "learning_rate": 5.846511212364872e-06,
      "loss": 0.0082,
      "step": 9165
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.3220323920249939,
      "learning_rate": 5.8375770570892535e-06,
      "loss": 0.0092,
      "step": 9170
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5357863903045654,
      "learning_rate": 5.828642901813634e-06,
      "loss": 0.0097,
      "step": 9175
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.6542167663574219,
      "learning_rate": 5.8197087465380155e-06,
      "loss": 0.0111,
      "step": 9180
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.3759448230266571,
      "learning_rate": 5.810774591262397e-06,
      "loss": 0.0076,
      "step": 9185
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.22233974933624268,
      "learning_rate": 5.801840435986778e-06,
      "loss": 0.0104,
      "step": 9190
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.3600682318210602,
      "learning_rate": 5.7929062807111595e-06,
      "loss": 0.008,
      "step": 9195
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.2026982307434082,
      "learning_rate": 5.7839721254355405e-06,
      "loss": 0.0143,
      "step": 9200
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.21646694839000702,
      "learning_rate": 5.7750379701599224e-06,
      "loss": 0.0117,
      "step": 9205
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.2780154347419739,
      "learning_rate": 5.766103814884303e-06,
      "loss": 0.0069,
      "step": 9210
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.38007307052612305,
      "learning_rate": 5.7571696596086845e-06,
      "loss": 0.0071,
      "step": 9215
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.23504765331745148,
      "learning_rate": 5.748235504333066e-06,
      "loss": 0.0073,
      "step": 9220
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.3849126100540161,
      "learning_rate": 5.739301349057447e-06,
      "loss": 0.0062,
      "step": 9225
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.39096078276634216,
      "learning_rate": 5.7303671937818285e-06,
      "loss": 0.0081,
      "step": 9230
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.35881391167640686,
      "learning_rate": 5.72143303850621e-06,
      "loss": 0.0073,
      "step": 9235
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.2720240652561188,
      "learning_rate": 5.7124988832305905e-06,
      "loss": 0.006,
      "step": 9240
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.17227132618427277,
      "learning_rate": 5.7035647279549724e-06,
      "loss": 0.0073,
      "step": 9245
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.4603612422943115,
      "learning_rate": 5.6946305726793535e-06,
      "loss": 0.0116,
      "step": 9250
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.6847078204154968,
      "learning_rate": 5.685696417403735e-06,
      "loss": 0.0075,
      "step": 9255
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.41257572174072266,
      "learning_rate": 5.676762262128116e-06,
      "loss": 0.0103,
      "step": 9260
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.402539998292923,
      "learning_rate": 5.6678281068524974e-06,
      "loss": 0.0063,
      "step": 9265
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.4992349147796631,
      "learning_rate": 5.658893951576879e-06,
      "loss": 0.0103,
      "step": 9270
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.47921329736709595,
      "learning_rate": 5.6499597963012595e-06,
      "loss": 0.0082,
      "step": 9275
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.2835437059402466,
      "learning_rate": 5.641025641025641e-06,
      "loss": 0.0085,
      "step": 9280
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.23391133546829224,
      "learning_rate": 5.632091485750023e-06,
      "loss": 0.0078,
      "step": 9285
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.3029947876930237,
      "learning_rate": 5.6231573304744035e-06,
      "loss": 0.0145,
      "step": 9290
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.26060959696769714,
      "learning_rate": 5.614223175198785e-06,
      "loss": 0.0061,
      "step": 9295
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.2901848554611206,
      "learning_rate": 5.605289019923167e-06,
      "loss": 0.0083,
      "step": 9300
    },
    {
      "epoch": 0.75,
      "eval_loss": 0.009584016166627407,
      "eval_pearson_cosine": 0.9615300255520981,
      "eval_pearson_dot": 0.9604189728759469,
      "eval_pearson_euclidean": 0.9517997399844298,
      "eval_pearson_manhattan": 0.9500464009872014,
      "eval_pearson_max": 0.9615300255520981,
      "eval_runtime": 425.5009,
      "eval_samples_per_second": 1.175,
      "eval_spearman_cosine": 0.9606140184560736,
      "eval_spearman_dot": 0.9573194772779091,
      "eval_spearman_euclidean": 0.9593890535562142,
      "eval_spearman_manhattan": 0.9579064236256946,
      "eval_spearman_max": 0.9606140184560736,
      "eval_steps_per_second": 1.175,
      "step": 9300
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.4470210373401642,
      "learning_rate": 5.596354864647548e-06,
      "loss": 0.0129,
      "step": 9305
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.4185398519039154,
      "learning_rate": 5.587420709371929e-06,
      "loss": 0.0081,
      "step": 9310
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.6620985269546509,
      "learning_rate": 5.57848655409631e-06,
      "loss": 0.0092,
      "step": 9315
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5238575339317322,
      "learning_rate": 5.569552398820692e-06,
      "loss": 0.0106,
      "step": 9320
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.8626998662948608,
      "learning_rate": 5.5606182435450724e-06,
      "loss": 0.0149,
      "step": 9325
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.312152624130249,
      "learning_rate": 5.551684088269454e-06,
      "loss": 0.0085,
      "step": 9330
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.2844025194644928,
      "learning_rate": 5.542749932993836e-06,
      "loss": 0.01,
      "step": 9335
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.4623136520385742,
      "learning_rate": 5.533815777718216e-06,
      "loss": 0.0082,
      "step": 9340
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.34604260325431824,
      "learning_rate": 5.524881622442598e-06,
      "loss": 0.0105,
      "step": 9345
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.5630137920379639,
      "learning_rate": 5.51594746716698e-06,
      "loss": 0.008,
      "step": 9350
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.22408334910869598,
      "learning_rate": 5.507013311891361e-06,
      "loss": 0.0086,
      "step": 9355
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.3562946915626526,
      "learning_rate": 5.498079156615742e-06,
      "loss": 0.0077,
      "step": 9360
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.6519134640693665,
      "learning_rate": 5.489145001340124e-06,
      "loss": 0.0273,
      "step": 9365
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.3609578609466553,
      "learning_rate": 5.480210846064505e-06,
      "loss": 0.008,
      "step": 9370
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.4362548291683197,
      "learning_rate": 5.471276690788886e-06,
      "loss": 0.0156,
      "step": 9375
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.51788729429245,
      "learning_rate": 5.462342535513267e-06,
      "loss": 0.0143,
      "step": 9380
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.3049197793006897,
      "learning_rate": 5.453408380237649e-06,
      "loss": 0.0084,
      "step": 9385
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.27513086795806885,
      "learning_rate": 5.444474224962031e-06,
      "loss": 0.0089,
      "step": 9390
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5995715260505676,
      "learning_rate": 5.435540069686411e-06,
      "loss": 0.0176,
      "step": 9395
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.4572034180164337,
      "learning_rate": 5.426605914410793e-06,
      "loss": 0.0079,
      "step": 9400
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.38861045241355896,
      "learning_rate": 5.417671759135175e-06,
      "loss": 0.0093,
      "step": 9405
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.2802937924861908,
      "learning_rate": 5.408737603859555e-06,
      "loss": 0.0099,
      "step": 9410
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.2714308500289917,
      "learning_rate": 5.399803448583937e-06,
      "loss": 0.0073,
      "step": 9415
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.4080353379249573,
      "learning_rate": 5.390869293308318e-06,
      "loss": 0.0069,
      "step": 9420
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.22111305594444275,
      "learning_rate": 5.381935138032699e-06,
      "loss": 0.0058,
      "step": 9425
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.7781515717506409,
      "learning_rate": 5.37300098275708e-06,
      "loss": 0.0066,
      "step": 9430
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.3610839545726776,
      "learning_rate": 5.364066827481462e-06,
      "loss": 0.0088,
      "step": 9435
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.3884970247745514,
      "learning_rate": 5.355132672205844e-06,
      "loss": 0.0094,
      "step": 9440
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5556257963180542,
      "learning_rate": 5.346198516930224e-06,
      "loss": 0.0105,
      "step": 9445
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5632768273353577,
      "learning_rate": 5.337264361654606e-06,
      "loss": 0.0091,
      "step": 9450
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.6423171758651733,
      "learning_rate": 5.328330206378988e-06,
      "loss": 0.0079,
      "step": 9455
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5519477725028992,
      "learning_rate": 5.319396051103368e-06,
      "loss": 0.0192,
      "step": 9460
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5822392106056213,
      "learning_rate": 5.31046189582775e-06,
      "loss": 0.0123,
      "step": 9465
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.49340376257896423,
      "learning_rate": 5.301527740552132e-06,
      "loss": 0.0083,
      "step": 9470
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.3505455553531647,
      "learning_rate": 5.292593585276512e-06,
      "loss": 0.011,
      "step": 9475
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.7089307308197021,
      "learning_rate": 5.283659430000894e-06,
      "loss": 0.0079,
      "step": 9480
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.791226863861084,
      "learning_rate": 5.274725274725275e-06,
      "loss": 0.0124,
      "step": 9485
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.38624152541160583,
      "learning_rate": 5.265791119449657e-06,
      "loss": 0.0076,
      "step": 9490
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.2802993953227997,
      "learning_rate": 5.256856964174037e-06,
      "loss": 0.0183,
      "step": 9495
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.4529344439506531,
      "learning_rate": 5.247922808898419e-06,
      "loss": 0.0112,
      "step": 9500
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.167479395866394,
      "learning_rate": 5.238988653622801e-06,
      "loss": 0.0128,
      "step": 9505
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.2423386126756668,
      "learning_rate": 5.230054498347181e-06,
      "loss": 0.0093,
      "step": 9510
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.3792467713356018,
      "learning_rate": 5.221120343071563e-06,
      "loss": 0.0078,
      "step": 9515
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.47463178634643555,
      "learning_rate": 5.212186187795945e-06,
      "loss": 0.0116,
      "step": 9520
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.9293251037597656,
      "learning_rate": 5.203252032520326e-06,
      "loss": 0.0107,
      "step": 9525
    },
    {
      "epoch": 0.77,
      "grad_norm": 41.848262786865234,
      "learning_rate": 5.194317877244707e-06,
      "loss": 0.0182,
      "step": 9530
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.34961816668510437,
      "learning_rate": 5.185383721969088e-06,
      "loss": 0.007,
      "step": 9535
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.822770595550537,
      "learning_rate": 5.17644956669347e-06,
      "loss": 0.0151,
      "step": 9540
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.3188020884990692,
      "learning_rate": 5.167515411417851e-06,
      "loss": 0.0067,
      "step": 9545
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.43374982476234436,
      "learning_rate": 5.158581256142232e-06,
      "loss": 0.0084,
      "step": 9550
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.4171868562698364,
      "learning_rate": 5.149647100866614e-06,
      "loss": 0.0073,
      "step": 9555
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5583686232566833,
      "learning_rate": 5.140712945590994e-06,
      "loss": 0.0092,
      "step": 9560
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.24175512790679932,
      "learning_rate": 5.131778790315376e-06,
      "loss": 0.0081,
      "step": 9565
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.3415161371231079,
      "learning_rate": 5.122844635039758e-06,
      "loss": 0.0109,
      "step": 9570
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5684280395507812,
      "learning_rate": 5.11391047976414e-06,
      "loss": 0.0072,
      "step": 9575
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.9785246849060059,
      "learning_rate": 5.10497632448852e-06,
      "loss": 0.0098,
      "step": 9580
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.2952157258987427,
      "learning_rate": 5.096042169212902e-06,
      "loss": 0.008,
      "step": 9585
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.20075848698616028,
      "learning_rate": 5.087108013937283e-06,
      "loss": 0.0074,
      "step": 9590
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.3542312681674957,
      "learning_rate": 5.078173858661664e-06,
      "loss": 0.0109,
      "step": 9595
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.38026684522628784,
      "learning_rate": 5.069239703386045e-06,
      "loss": 0.0069,
      "step": 9600
    },
    {
      "epoch": 0.77,
      "eval_loss": 0.008200183510780334,
      "eval_pearson_cosine": 0.9642878383068713,
      "eval_pearson_dot": 0.9627028886298191,
      "eval_pearson_euclidean": 0.9523746354427236,
      "eval_pearson_manhattan": 0.9507404567548948,
      "eval_pearson_max": 0.9642878383068713,
      "eval_runtime": 425.6137,
      "eval_samples_per_second": 1.175,
      "eval_spearman_cosine": 0.9605806103224412,
      "eval_spearman_dot": 0.955501806007224,
      "eval_spearman_euclidean": 0.9597567350269401,
      "eval_spearman_manhattan": 0.9581544886179545,
      "eval_spearman_max": 0.9605806103224412,
      "eval_steps_per_second": 1.175,
      "step": 9600
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.37074190378189087,
      "learning_rate": 5.060305548110427e-06,
      "loss": 0.0173,
      "step": 9605
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.3430071175098419,
      "learning_rate": 5.051371392834808e-06,
      "loss": 0.0077,
      "step": 9610
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.516249418258667,
      "learning_rate": 5.042437237559189e-06,
      "loss": 0.0059,
      "step": 9615
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.3532971143722534,
      "learning_rate": 5.033503082283571e-06,
      "loss": 0.0113,
      "step": 9620
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.4686454236507416,
      "learning_rate": 5.0245689270079526e-06,
      "loss": 0.0097,
      "step": 9625
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.2624160051345825,
      "learning_rate": 5.015634771732333e-06,
      "loss": 0.0087,
      "step": 9630
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.35835760831832886,
      "learning_rate": 5.006700616456715e-06,
      "loss": 0.0076,
      "step": 9635
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.6053426265716553,
      "learning_rate": 4.997766461181096e-06,
      "loss": 0.0158,
      "step": 9640
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.399104505777359,
      "learning_rate": 4.9888323059054776e-06,
      "loss": 0.0085,
      "step": 9645
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.3674059212207794,
      "learning_rate": 4.979898150629859e-06,
      "loss": 0.0112,
      "step": 9650
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.16110941767692566,
      "learning_rate": 4.97096399535424e-06,
      "loss": 0.0086,
      "step": 9655
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.2616519331932068,
      "learning_rate": 4.962029840078621e-06,
      "loss": 0.009,
      "step": 9660
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.6079025268554688,
      "learning_rate": 4.953095684803002e-06,
      "loss": 0.0082,
      "step": 9665
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.30316200852394104,
      "learning_rate": 4.944161529527384e-06,
      "loss": 0.0063,
      "step": 9670
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.36754128336906433,
      "learning_rate": 4.935227374251765e-06,
      "loss": 0.0073,
      "step": 9675
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.6272875666618347,
      "learning_rate": 4.9262932189761465e-06,
      "loss": 0.0089,
      "step": 9680
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.5529404878616333,
      "learning_rate": 4.9173590637005276e-06,
      "loss": 0.0095,
      "step": 9685
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.2829398214817047,
      "learning_rate": 4.908424908424909e-06,
      "loss": 0.011,
      "step": 9690
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.30931296944618225,
      "learning_rate": 4.8994907531492905e-06,
      "loss": 0.0103,
      "step": 9695
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.6834177374839783,
      "learning_rate": 4.8905565978736715e-06,
      "loss": 0.0087,
      "step": 9700
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.42869821190834045,
      "learning_rate": 4.8816224425980526e-06,
      "loss": 0.0088,
      "step": 9705
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.3291241228580475,
      "learning_rate": 4.8726882873224344e-06,
      "loss": 0.0061,
      "step": 9710
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.3477993607521057,
      "learning_rate": 4.8637541320468155e-06,
      "loss": 0.007,
      "step": 9715
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.393031507730484,
      "learning_rate": 4.8548199767711965e-06,
      "loss": 0.0077,
      "step": 9720
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.37427353858947754,
      "learning_rate": 4.8458858214955776e-06,
      "loss": 0.0086,
      "step": 9725
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.4370558261871338,
      "learning_rate": 4.8369516662199594e-06,
      "loss": 0.0064,
      "step": 9730
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.31545019149780273,
      "learning_rate": 4.8280175109443405e-06,
      "loss": 0.0059,
      "step": 9735
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.7945960760116577,
      "learning_rate": 4.8190833556687215e-06,
      "loss": 0.021,
      "step": 9740
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.29888418316841125,
      "learning_rate": 4.810149200393103e-06,
      "loss": 0.0062,
      "step": 9745
    },
    {
      "epoch": 0.78,
      "grad_norm": 3.3094396591186523,
      "learning_rate": 4.8012150451174844e-06,
      "loss": 0.0154,
      "step": 9750
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.2018340677022934,
      "learning_rate": 4.792280889841866e-06,
      "loss": 0.0074,
      "step": 9755
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.7358143329620361,
      "learning_rate": 4.783346734566247e-06,
      "loss": 0.0154,
      "step": 9760
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.580872654914856,
      "learning_rate": 4.774412579290628e-06,
      "loss": 0.0111,
      "step": 9765
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.3042278289794922,
      "learning_rate": 4.7654784240150095e-06,
      "loss": 0.0078,
      "step": 9770
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.42176923155784607,
      "learning_rate": 4.7565442687393905e-06,
      "loss": 0.0062,
      "step": 9775
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.30140987038612366,
      "learning_rate": 4.747610113463772e-06,
      "loss": 0.0063,
      "step": 9780
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5304137468338013,
      "learning_rate": 4.738675958188153e-06,
      "loss": 0.0097,
      "step": 9785
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.36996015906333923,
      "learning_rate": 4.729741802912535e-06,
      "loss": 0.0097,
      "step": 9790
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.6667109131813049,
      "learning_rate": 4.720807647636916e-06,
      "loss": 0.0075,
      "step": 9795
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.2762182950973511,
      "learning_rate": 4.711873492361297e-06,
      "loss": 0.0064,
      "step": 9800
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.3733229339122772,
      "learning_rate": 4.702939337085679e-06,
      "loss": 0.0054,
      "step": 9805
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.3811498284339905,
      "learning_rate": 4.69400518181006e-06,
      "loss": 0.0062,
      "step": 9810
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5551919937133789,
      "learning_rate": 4.685071026534442e-06,
      "loss": 0.0089,
      "step": 9815
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5745194554328918,
      "learning_rate": 4.676136871258823e-06,
      "loss": 0.0097,
      "step": 9820
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.2763228118419647,
      "learning_rate": 4.667202715983204e-06,
      "loss": 0.01,
      "step": 9825
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5403454899787903,
      "learning_rate": 4.658268560707585e-06,
      "loss": 0.0134,
      "step": 9830
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.37194764614105225,
      "learning_rate": 4.649334405431966e-06,
      "loss": 0.0051,
      "step": 9835
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.21742063760757446,
      "learning_rate": 4.640400250156348e-06,
      "loss": 0.0123,
      "step": 9840
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.3551539182662964,
      "learning_rate": 4.631466094880729e-06,
      "loss": 0.0104,
      "step": 9845
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.248150035738945,
      "learning_rate": 4.62253193960511e-06,
      "loss": 0.0054,
      "step": 9850
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.6017441153526306,
      "learning_rate": 4.613597784329492e-06,
      "loss": 0.0115,
      "step": 9855
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.46963444352149963,
      "learning_rate": 4.604663629053873e-06,
      "loss": 0.0087,
      "step": 9860
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.26475605368614197,
      "learning_rate": 4.595729473778255e-06,
      "loss": 0.0064,
      "step": 9865
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.3036366403102875,
      "learning_rate": 4.586795318502636e-06,
      "loss": 0.0094,
      "step": 9870
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.4167456328868866,
      "learning_rate": 4.577861163227017e-06,
      "loss": 0.008,
      "step": 9875
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.27215877175331116,
      "learning_rate": 4.568927007951398e-06,
      "loss": 0.0087,
      "step": 9880
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.3947705924510956,
      "learning_rate": 4.55999285267578e-06,
      "loss": 0.0099,
      "step": 9885
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.261850506067276,
      "learning_rate": 4.551058697400161e-06,
      "loss": 0.012,
      "step": 9890
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.27852803468704224,
      "learning_rate": 4.542124542124542e-06,
      "loss": 0.0063,
      "step": 9895
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.3846147060394287,
      "learning_rate": 4.533190386848923e-06,
      "loss": 0.0072,
      "step": 9900
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.007968730293214321,
      "eval_pearson_cosine": 0.9658888665742654,
      "eval_pearson_dot": 0.9627392628207877,
      "eval_pearson_euclidean": 0.9531911193892233,
      "eval_pearson_manhattan": 0.9513587882321657,
      "eval_pearson_max": 0.9658888665742654,
      "eval_runtime": 425.901,
      "eval_samples_per_second": 1.174,
      "eval_spearman_cosine": 0.9624986019944078,
      "eval_spearman_dot": 0.955504398017592,
      "eval_spearman_euclidean": 0.9630360121440484,
      "eval_spearman_manhattan": 0.9610320041280165,
      "eval_spearman_max": 0.9630360121440484,
      "eval_steps_per_second": 1.174,
      "step": 9900
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.45619475841522217,
      "learning_rate": 4.524256231573305e-06,
      "loss": 0.009,
      "step": 9905
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.39344677329063416,
      "learning_rate": 4.515322076297686e-06,
      "loss": 0.0088,
      "step": 9910
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.18160255253314972,
      "learning_rate": 4.506387921022068e-06,
      "loss": 0.0059,
      "step": 9915
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.19619829952716827,
      "learning_rate": 4.497453765746449e-06,
      "loss": 0.0081,
      "step": 9920
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.2846349775791168,
      "learning_rate": 4.488519610470831e-06,
      "loss": 0.0049,
      "step": 9925
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.3724232316017151,
      "learning_rate": 4.479585455195212e-06,
      "loss": 0.0118,
      "step": 9930
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.3385705351829529,
      "learning_rate": 4.470651299919593e-06,
      "loss": 0.0069,
      "step": 9935
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.2874641418457031,
      "learning_rate": 4.461717144643974e-06,
      "loss": 0.006,
      "step": 9940
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.3358646035194397,
      "learning_rate": 4.452782989368355e-06,
      "loss": 0.0062,
      "step": 9945
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.7037550806999207,
      "learning_rate": 4.443848834092737e-06,
      "loss": 0.009,
      "step": 9950
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.20332568883895874,
      "learning_rate": 4.434914678817118e-06,
      "loss": 0.0084,
      "step": 9955
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.36842986941337585,
      "learning_rate": 4.425980523541499e-06,
      "loss": 0.0087,
      "step": 9960
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.283997505903244,
      "learning_rate": 4.417046368265881e-06,
      "loss": 0.0086,
      "step": 9965
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.42346441745758057,
      "learning_rate": 4.408112212990262e-06,
      "loss": 0.0078,
      "step": 9970
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.18222256004810333,
      "learning_rate": 4.399178057714644e-06,
      "loss": 0.0074,
      "step": 9975
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.6875673532485962,
      "learning_rate": 4.390243902439025e-06,
      "loss": 0.0102,
      "step": 9980
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.24472391605377197,
      "learning_rate": 4.381309747163406e-06,
      "loss": 0.0065,
      "step": 9985
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.26704928278923035,
      "learning_rate": 4.372375591887788e-06,
      "loss": 0.0083,
      "step": 9990
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5033184289932251,
      "learning_rate": 4.363441436612169e-06,
      "loss": 0.0067,
      "step": 9995
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.781326413154602,
      "learning_rate": 4.35450728133655e-06,
      "loss": 0.006,
      "step": 10000
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.46775344014167786,
      "learning_rate": 4.345573126060931e-06,
      "loss": 0.0087,
      "step": 10005
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.3803477883338928,
      "learning_rate": 4.336638970785312e-06,
      "loss": 0.0098,
      "step": 10010
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.23086823523044586,
      "learning_rate": 4.327704815509694e-06,
      "loss": 0.0089,
      "step": 10015
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.18261872231960297,
      "learning_rate": 4.318770660234075e-06,
      "loss": 0.01,
      "step": 10020
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.33674633502960205,
      "learning_rate": 4.309836504958457e-06,
      "loss": 0.0086,
      "step": 10025
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.29859867691993713,
      "learning_rate": 4.300902349682838e-06,
      "loss": 0.009,
      "step": 10030
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.40897712111473083,
      "learning_rate": 4.291968194407219e-06,
      "loss": 0.0092,
      "step": 10035
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.7651856541633606,
      "learning_rate": 4.283034039131601e-06,
      "loss": 0.0108,
      "step": 10040
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.7065618634223938,
      "learning_rate": 4.274099883855982e-06,
      "loss": 0.0132,
      "step": 10045
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.4190121293067932,
      "learning_rate": 4.265165728580363e-06,
      "loss": 0.0063,
      "step": 10050
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.47081393003463745,
      "learning_rate": 4.256231573304745e-06,
      "loss": 0.0065,
      "step": 10055
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.3828545808792114,
      "learning_rate": 4.247297418029126e-06,
      "loss": 0.0087,
      "step": 10060
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.7525375485420227,
      "learning_rate": 4.238363262753507e-06,
      "loss": 0.0113,
      "step": 10065
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.34589239954948425,
      "learning_rate": 4.229429107477888e-06,
      "loss": 0.0078,
      "step": 10070
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5909443497657776,
      "learning_rate": 4.22049495220227e-06,
      "loss": 0.007,
      "step": 10075
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.3668850362300873,
      "learning_rate": 4.211560796926651e-06,
      "loss": 0.0061,
      "step": 10080
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.4989503026008606,
      "learning_rate": 4.202626641651033e-06,
      "loss": 0.0071,
      "step": 10085
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.44953587651252747,
      "learning_rate": 4.193692486375414e-06,
      "loss": 0.0079,
      "step": 10090
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.36224547028541565,
      "learning_rate": 4.184758331099795e-06,
      "loss": 0.0101,
      "step": 10095
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.21505020558834076,
      "learning_rate": 4.175824175824177e-06,
      "loss": 0.0116,
      "step": 10100
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.5628384351730347,
      "learning_rate": 4.166890020548558e-06,
      "loss": 0.0106,
      "step": 10105
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.6173145174980164,
      "learning_rate": 4.157955865272939e-06,
      "loss": 0.0074,
      "step": 10110
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.20209026336669922,
      "learning_rate": 4.14902170999732e-06,
      "loss": 0.0054,
      "step": 10115
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.26845335960388184,
      "learning_rate": 4.140087554721701e-06,
      "loss": 0.0084,
      "step": 10120
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.19294553995132446,
      "learning_rate": 4.131153399446083e-06,
      "loss": 0.0069,
      "step": 10125
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.2686295211315155,
      "learning_rate": 4.122219244170464e-06,
      "loss": 0.0062,
      "step": 10130
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.34871765971183777,
      "learning_rate": 4.113285088894846e-06,
      "loss": 0.0087,
      "step": 10135
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.3453786373138428,
      "learning_rate": 4.104350933619227e-06,
      "loss": 0.008,
      "step": 10140
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.4701385200023651,
      "learning_rate": 4.095416778343608e-06,
      "loss": 0.0077,
      "step": 10145
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.13584518432617188,
      "learning_rate": 4.0864826230679896e-06,
      "loss": 0.0063,
      "step": 10150
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5030553936958313,
      "learning_rate": 4.077548467792371e-06,
      "loss": 0.0076,
      "step": 10155
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.44598788022994995,
      "learning_rate": 4.0686143125167525e-06,
      "loss": 0.0067,
      "step": 10160
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.2886448800563812,
      "learning_rate": 4.0596801572411335e-06,
      "loss": 0.0077,
      "step": 10165
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.2822360098361969,
      "learning_rate": 4.0507460019655146e-06,
      "loss": 0.0078,
      "step": 10170
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5592710375785828,
      "learning_rate": 4.041811846689896e-06,
      "loss": 0.0076,
      "step": 10175
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.310092031955719,
      "learning_rate": 4.032877691414277e-06,
      "loss": 0.0107,
      "step": 10180
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.3115810751914978,
      "learning_rate": 4.0239435361386585e-06,
      "loss": 0.0087,
      "step": 10185
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.9056434035301208,
      "learning_rate": 4.0150093808630396e-06,
      "loss": 0.0099,
      "step": 10190
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.31844180822372437,
      "learning_rate": 4.006075225587421e-06,
      "loss": 0.0111,
      "step": 10195
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5329269766807556,
      "learning_rate": 3.9971410703118025e-06,
      "loss": 0.01,
      "step": 10200
    },
    {
      "epoch": 0.82,
      "eval_loss": 0.007998097687959671,
      "eval_pearson_cosine": 0.9672417410329249,
      "eval_pearson_dot": 0.965013249775285,
      "eval_pearson_euclidean": 0.9542839853098279,
      "eval_pearson_manhattan": 0.9528819643748915,
      "eval_pearson_max": 0.9672417410329249,
      "eval_runtime": 425.8541,
      "eval_samples_per_second": 1.174,
      "eval_spearman_cosine": 0.9626618986475944,
      "eval_spearman_dot": 0.9577223908895635,
      "eval_spearman_euclidean": 0.962258313033252,
      "eval_spearman_manhattan": 0.9619932559730238,
      "eval_spearman_max": 0.9626618986475944,
      "eval_steps_per_second": 1.174,
      "step": 10200
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.3950428068637848,
      "learning_rate": 3.9882069150361835e-06,
      "loss": 0.0074,
      "step": 10205
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.3376205563545227,
      "learning_rate": 3.979272759760565e-06,
      "loss": 0.0079,
      "step": 10210
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.39471399784088135,
      "learning_rate": 3.9703386044849465e-06,
      "loss": 0.0087,
      "step": 10215
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.42924797534942627,
      "learning_rate": 3.9614044492093275e-06,
      "loss": 0.0086,
      "step": 10220
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.26988697052001953,
      "learning_rate": 3.9524702939337085e-06,
      "loss": 0.0076,
      "step": 10225
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5494524240493774,
      "learning_rate": 3.94353613865809e-06,
      "loss": 0.0096,
      "step": 10230
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.39503470063209534,
      "learning_rate": 3.9346019833824715e-06,
      "loss": 0.0091,
      "step": 10235
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.2849455177783966,
      "learning_rate": 3.9256678281068525e-06,
      "loss": 0.0066,
      "step": 10240
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.2523050904273987,
      "learning_rate": 3.916733672831234e-06,
      "loss": 0.0076,
      "step": 10245
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.4256332218647003,
      "learning_rate": 3.9077995175556154e-06,
      "loss": 0.0061,
      "step": 10250
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.2837385833263397,
      "learning_rate": 3.8988653622799965e-06,
      "loss": 0.0069,
      "step": 10255
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.44190511107444763,
      "learning_rate": 3.889931207004378e-06,
      "loss": 0.0059,
      "step": 10260
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.3654380440711975,
      "learning_rate": 3.880997051728759e-06,
      "loss": 0.0071,
      "step": 10265
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.356179416179657,
      "learning_rate": 3.872062896453141e-06,
      "loss": 0.005,
      "step": 10270
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.4459650218486786,
      "learning_rate": 3.863128741177522e-06,
      "loss": 0.0074,
      "step": 10275
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.6780635118484497,
      "learning_rate": 3.854194585901903e-06,
      "loss": 0.0101,
      "step": 10280
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.42910513281822205,
      "learning_rate": 3.845260430626284e-06,
      "loss": 0.0087,
      "step": 10285
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.38253623247146606,
      "learning_rate": 3.8363262753506654e-06,
      "loss": 0.0054,
      "step": 10290
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.3616214990615845,
      "learning_rate": 3.827392120075047e-06,
      "loss": 0.0066,
      "step": 10295
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5730588436126709,
      "learning_rate": 3.818457964799428e-06,
      "loss": 0.0075,
      "step": 10300
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.34027931094169617,
      "learning_rate": 3.80952380952381e-06,
      "loss": 0.0115,
      "step": 10305
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.34853848814964294,
      "learning_rate": 3.8005896542481913e-06,
      "loss": 0.0098,
      "step": 10310
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.3500560522079468,
      "learning_rate": 3.7916554989725723e-06,
      "loss": 0.0099,
      "step": 10315
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.4545835256576538,
      "learning_rate": 3.7827213436969538e-06,
      "loss": 0.0075,
      "step": 10320
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.3421791195869446,
      "learning_rate": 3.7737871884213352e-06,
      "loss": 0.0105,
      "step": 10325
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.28388506174087524,
      "learning_rate": 3.7648530331457163e-06,
      "loss": 0.0135,
      "step": 10330
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.6649767160415649,
      "learning_rate": 3.7559188778700977e-06,
      "loss": 0.0105,
      "step": 10335
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.41207408905029297,
      "learning_rate": 3.7469847225944788e-06,
      "loss": 0.0065,
      "step": 10340
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.3896176815032959,
      "learning_rate": 3.7380505673188607e-06,
      "loss": 0.0098,
      "step": 10345
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.26551979780197144,
      "learning_rate": 3.7291164120432417e-06,
      "loss": 0.0088,
      "step": 10350
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.22879204154014587,
      "learning_rate": 3.7201822567676227e-06,
      "loss": 0.0094,
      "step": 10355
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.6052958369255066,
      "learning_rate": 3.711248101492004e-06,
      "loss": 0.0092,
      "step": 10360
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.2694813311100006,
      "learning_rate": 3.7023139462163852e-06,
      "loss": 0.0054,
      "step": 10365
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.3190039098262787,
      "learning_rate": 3.693379790940767e-06,
      "loss": 0.009,
      "step": 10370
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.2353006899356842,
      "learning_rate": 3.684445635665148e-06,
      "loss": 0.0088,
      "step": 10375
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.9957902431488037,
      "learning_rate": 3.675511480389529e-06,
      "loss": 0.0116,
      "step": 10380
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.4154163897037506,
      "learning_rate": 3.6665773251139107e-06,
      "loss": 0.0055,
      "step": 10385
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.3367329239845276,
      "learning_rate": 3.657643169838292e-06,
      "loss": 0.0061,
      "step": 10390
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.3034825325012207,
      "learning_rate": 3.6487090145626736e-06,
      "loss": 0.0081,
      "step": 10395
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.17845579981803894,
      "learning_rate": 3.6397748592870546e-06,
      "loss": 0.0043,
      "step": 10400
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.4659731984138489,
      "learning_rate": 3.630840704011436e-06,
      "loss": 0.0064,
      "step": 10405
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.8237358331680298,
      "learning_rate": 3.6219065487358176e-06,
      "loss": 0.0089,
      "step": 10410
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.2651444971561432,
      "learning_rate": 3.6129723934601986e-06,
      "loss": 0.0088,
      "step": 10415
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.2811392545700073,
      "learning_rate": 3.60403823818458e-06,
      "loss": 0.0045,
      "step": 10420
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.26526081562042236,
      "learning_rate": 3.595104082908961e-06,
      "loss": 0.0077,
      "step": 10425
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.31172770261764526,
      "learning_rate": 3.586169927633343e-06,
      "loss": 0.0066,
      "step": 10430
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.24446207284927368,
      "learning_rate": 3.577235772357724e-06,
      "loss": 0.0095,
      "step": 10435
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.36921826004981995,
      "learning_rate": 3.568301617082105e-06,
      "loss": 0.0092,
      "step": 10440
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.37674766778945923,
      "learning_rate": 3.5593674618064865e-06,
      "loss": 0.0068,
      "step": 10445
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.3865291178226471,
      "learning_rate": 3.5504333065308676e-06,
      "loss": 0.0101,
      "step": 10450
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.27692753076553345,
      "learning_rate": 3.5414991512552494e-06,
      "loss": 0.0118,
      "step": 10455
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.3851732611656189,
      "learning_rate": 3.5325649959796305e-06,
      "loss": 0.0064,
      "step": 10460
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.3045642673969269,
      "learning_rate": 3.5236308407040115e-06,
      "loss": 0.006,
      "step": 10465
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.33417442440986633,
      "learning_rate": 3.514696685428393e-06,
      "loss": 0.0053,
      "step": 10470
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.25378555059432983,
      "learning_rate": 3.505762530152774e-06,
      "loss": 0.0054,
      "step": 10475
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.30482515692710876,
      "learning_rate": 3.496828374877156e-06,
      "loss": 0.006,
      "step": 10480
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.3786448836326599,
      "learning_rate": 3.487894219601537e-06,
      "loss": 0.0069,
      "step": 10485
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.4568743407726288,
      "learning_rate": 3.478960064325918e-06,
      "loss": 0.007,
      "step": 10490
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.4057961702346802,
      "learning_rate": 3.4700259090503e-06,
      "loss": 0.0072,
      "step": 10495
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.4396592080593109,
      "learning_rate": 3.461091753774681e-06,
      "loss": 0.0068,
      "step": 10500
    },
    {
      "epoch": 0.84,
      "eval_loss": 0.0074067204259335995,
      "eval_pearson_cosine": 0.9673687223458171,
      "eval_pearson_dot": 0.9653915362220337,
      "eval_pearson_euclidean": 0.9564555475438925,
      "eval_pearson_manhattan": 0.9551556309301102,
      "eval_pearson_max": 0.9673687223458171,
      "eval_runtime": 426.0441,
      "eval_samples_per_second": 1.174,
      "eval_spearman_cosine": 0.9609645158580634,
      "eval_spearman_dot": 0.955628814515258,
      "eval_spearman_euclidean": 0.9613639734558937,
      "eval_spearman_manhattan": 0.9601420805683222,
      "eval_spearman_max": 0.9613639734558937,
      "eval_steps_per_second": 1.174,
      "step": 10500
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.33185890316963196,
      "learning_rate": 3.4521575984990624e-06,
      "loss": 0.0108,
      "step": 10505
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.42153021693229675,
      "learning_rate": 3.4432234432234434e-06,
      "loss": 0.0111,
      "step": 10510
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.3933052122592926,
      "learning_rate": 3.4342892879478245e-06,
      "loss": 0.0053,
      "step": 10515
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.37213289737701416,
      "learning_rate": 3.4253551326722063e-06,
      "loss": 0.0052,
      "step": 10520
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.3832128643989563,
      "learning_rate": 3.4164209773965874e-06,
      "loss": 0.0083,
      "step": 10525
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.2269567996263504,
      "learning_rate": 3.407486822120969e-06,
      "loss": 0.0046,
      "step": 10530
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.3588921129703522,
      "learning_rate": 3.39855266684535e-06,
      "loss": 0.0085,
      "step": 10535
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.2792292833328247,
      "learning_rate": 3.389618511569731e-06,
      "loss": 0.0064,
      "step": 10540
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.30247360467910767,
      "learning_rate": 3.380684356294113e-06,
      "loss": 0.009,
      "step": 10545
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.33265256881713867,
      "learning_rate": 3.371750201018494e-06,
      "loss": 0.0087,
      "step": 10550
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.41412991285324097,
      "learning_rate": 3.3628160457428753e-06,
      "loss": 0.0059,
      "step": 10555
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.31760454177856445,
      "learning_rate": 3.3538818904672563e-06,
      "loss": 0.0062,
      "step": 10560
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.3177451193332672,
      "learning_rate": 3.3449477351916382e-06,
      "loss": 0.0072,
      "step": 10565
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.45792147517204285,
      "learning_rate": 3.3360135799160193e-06,
      "loss": 0.0075,
      "step": 10570
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.29566365480422974,
      "learning_rate": 3.3270794246404003e-06,
      "loss": 0.0059,
      "step": 10575
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.32841619849205017,
      "learning_rate": 3.318145269364782e-06,
      "loss": 0.0108,
      "step": 10580
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.341621458530426,
      "learning_rate": 3.3092111140891632e-06,
      "loss": 0.0128,
      "step": 10585
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.423700213432312,
      "learning_rate": 3.3002769588135447e-06,
      "loss": 0.0075,
      "step": 10590
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.3334985673427582,
      "learning_rate": 3.2913428035379257e-06,
      "loss": 0.0085,
      "step": 10595
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.4427613317966461,
      "learning_rate": 3.2824086482623068e-06,
      "loss": 0.0098,
      "step": 10600
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.510867714881897,
      "learning_rate": 3.2734744929866887e-06,
      "loss": 0.007,
      "step": 10605
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.2945081293582916,
      "learning_rate": 3.2645403377110697e-06,
      "loss": 0.0095,
      "step": 10610
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.3742575943470001,
      "learning_rate": 3.255606182435451e-06,
      "loss": 0.0052,
      "step": 10615
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.3119674026966095,
      "learning_rate": 3.246672027159832e-06,
      "loss": 0.009,
      "step": 10620
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.2969549894332886,
      "learning_rate": 3.2377378718842132e-06,
      "loss": 0.007,
      "step": 10625
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.3154788315296173,
      "learning_rate": 3.228803716608595e-06,
      "loss": 0.0083,
      "step": 10630
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.33745089173316956,
      "learning_rate": 3.219869561332976e-06,
      "loss": 0.0056,
      "step": 10635
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.36266443133354187,
      "learning_rate": 3.2109354060573576e-06,
      "loss": 0.0087,
      "step": 10640
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.40266337990760803,
      "learning_rate": 3.2020012507817387e-06,
      "loss": 0.0055,
      "step": 10645
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.3595188558101654,
      "learning_rate": 3.19306709550612e-06,
      "loss": 0.0199,
      "step": 10650
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.3521510064601898,
      "learning_rate": 3.1841329402305016e-06,
      "loss": 0.0082,
      "step": 10655
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.3168518841266632,
      "learning_rate": 3.1751987849548826e-06,
      "loss": 0.009,
      "step": 10660
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4278966188430786,
      "learning_rate": 3.166264629679264e-06,
      "loss": 0.0075,
      "step": 10665
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.379189670085907,
      "learning_rate": 3.1573304744036455e-06,
      "loss": 0.008,
      "step": 10670
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.361432820558548,
      "learning_rate": 3.1483963191280266e-06,
      "loss": 0.0069,
      "step": 10675
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.20192596316337585,
      "learning_rate": 3.139462163852408e-06,
      "loss": 0.0049,
      "step": 10680
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4057570695877075,
      "learning_rate": 3.130528008576789e-06,
      "loss": 0.0058,
      "step": 10685
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.2621855139732361,
      "learning_rate": 3.121593853301171e-06,
      "loss": 0.007,
      "step": 10690
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4035142958164215,
      "learning_rate": 3.112659698025552e-06,
      "loss": 0.0077,
      "step": 10695
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.35792276263237,
      "learning_rate": 3.103725542749933e-06,
      "loss": 0.0127,
      "step": 10700
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.48556092381477356,
      "learning_rate": 3.0947913874743145e-06,
      "loss": 0.008,
      "step": 10705
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.33445674180984497,
      "learning_rate": 3.0858572321986955e-06,
      "loss": 0.0059,
      "step": 10710
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4826265871524811,
      "learning_rate": 3.0769230769230774e-06,
      "loss": 0.0089,
      "step": 10715
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.42223629355430603,
      "learning_rate": 3.0679889216474585e-06,
      "loss": 0.0066,
      "step": 10720
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.45981764793395996,
      "learning_rate": 3.05905476637184e-06,
      "loss": 0.0085,
      "step": 10725
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.2595252990722656,
      "learning_rate": 3.050120611096221e-06,
      "loss": 0.0063,
      "step": 10730
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5080291032791138,
      "learning_rate": 3.0411864558206024e-06,
      "loss": 0.0094,
      "step": 10735
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.32294949889183044,
      "learning_rate": 3.032252300544984e-06,
      "loss": 0.0058,
      "step": 10740
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4228246808052063,
      "learning_rate": 3.023318145269365e-06,
      "loss": 0.0059,
      "step": 10745
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.3359188139438629,
      "learning_rate": 3.0143839899937464e-06,
      "loss": 0.0062,
      "step": 10750
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.2471199929714203,
      "learning_rate": 3.005449834718128e-06,
      "loss": 0.0066,
      "step": 10755
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.8795719742774963,
      "learning_rate": 2.996515679442509e-06,
      "loss": 0.0129,
      "step": 10760
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.3051950931549072,
      "learning_rate": 2.9875815241668904e-06,
      "loss": 0.0128,
      "step": 10765
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.3568212389945984,
      "learning_rate": 2.9786473688912714e-06,
      "loss": 0.0061,
      "step": 10770
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.3010600805282593,
      "learning_rate": 2.9697132136156533e-06,
      "loss": 0.0065,
      "step": 10775
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.38424891233444214,
      "learning_rate": 2.9607790583400343e-06,
      "loss": 0.0078,
      "step": 10780
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.309994101524353,
      "learning_rate": 2.9518449030644154e-06,
      "loss": 0.0086,
      "step": 10785
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.37481045722961426,
      "learning_rate": 2.942910747788797e-06,
      "loss": 0.0122,
      "step": 10790
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.432425856590271,
      "learning_rate": 2.933976592513178e-06,
      "loss": 0.0065,
      "step": 10795
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.27843984961509705,
      "learning_rate": 2.9250424372375598e-06,
      "loss": 0.0076,
      "step": 10800
    },
    {
      "epoch": 0.87,
      "eval_loss": 0.007469375152140856,
      "eval_pearson_cosine": 0.9681370387245068,
      "eval_pearson_dot": 0.9657977526847813,
      "eval_pearson_euclidean": 0.957258084356729,
      "eval_pearson_manhattan": 0.956023768146456,
      "eval_pearson_max": 0.9681370387245068,
      "eval_runtime": 426.2283,
      "eval_samples_per_second": 1.173,
      "eval_spearman_cosine": 0.9642074088296352,
      "eval_spearman_dot": 0.9586846987387948,
      "eval_spearman_euclidean": 0.9637806871227486,
      "eval_spearman_manhattan": 0.9624153696614787,
      "eval_spearman_max": 0.9642074088296352,
      "eval_steps_per_second": 1.173,
      "step": 10800
    }
  ],
  "logging_steps": 5,
  "max_steps": 12437,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 400,
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}