{
  "best_metric": 0.7912728190422058,
  "best_model_checkpoint": "saves/starcoder2-7b/lora/sft/checkpoint-1000",
  "epoch": 0.1596169193934557,
  "eval_steps": 100,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 0.628385603427887,
      "learning_rate": 4.999999126897802e-05,
      "loss": 1.2582,
      "step": 5
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.0855119228363037,
      "learning_rate": 4.999996507591817e-05,
      "loss": 0.801,
      "step": 10
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.5689586400985718,
      "learning_rate": 4.9999921420838745e-05,
      "loss": 1.067,
      "step": 15
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.0851330757141113,
      "learning_rate": 4.999986030377024e-05,
      "loss": 1.2953,
      "step": 20
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.397479772567749,
      "learning_rate": 4.999978172475535e-05,
      "loss": 0.9826,
      "step": 25
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.344118595123291,
      "learning_rate": 4.9999685683848954e-05,
      "loss": 0.9485,
      "step": 30
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.158163070678711,
      "learning_rate": 4.9999596278606616e-05,
      "loss": 0.8103,
      "step": 35
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.602233648300171,
      "learning_rate": 4.999946880647276e-05,
      "loss": 0.8648,
      "step": 40
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.557242751121521,
      "learning_rate": 4.999932387266596e-05,
      "loss": 1.0198,
      "step": 45
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.36068856716156,
      "learning_rate": 4.999916147728746e-05,
      "loss": 0.9367,
      "step": 50
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.3263639211654663,
      "learning_rate": 4.999898162045068e-05,
      "loss": 0.9695,
      "step": 55
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.333601474761963,
      "learning_rate": 4.999878430228126e-05,
      "loss": 1.1509,
      "step": 60
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.4753800630569458,
      "learning_rate": 4.999856952291702e-05,
      "loss": 1.1461,
      "step": 65
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.5096240043640137,
      "learning_rate": 4.9998337282507965e-05,
      "loss": 1.1722,
      "step": 70
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.189892053604126,
      "learning_rate": 4.999808758121633e-05,
      "loss": 1.1834,
      "step": 75
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.9292634725570679,
      "learning_rate": 4.999782041921651e-05,
      "loss": 0.9498,
      "step": 80
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.1775777339935303,
      "learning_rate": 4.9997535796695134e-05,
      "loss": 0.9346,
      "step": 85
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.6854296922683716,
      "learning_rate": 4.999723371385099e-05,
      "loss": 1.119,
      "step": 90
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.4571490287780762,
      "learning_rate": 4.999691417089507e-05,
      "loss": 0.8671,
      "step": 95
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.277044653892517,
      "learning_rate": 4.999657716805059e-05,
      "loss": 1.2469,
      "step": 100
    },
    {
      "epoch": 0.02,
      "eval_loss": 0.8478816747665405,
      "eval_runtime": 96.2736,
      "eval_samples_per_second": 7.24,
      "eval_steps_per_second": 7.24,
      "step": 100
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.6687743067741394,
      "learning_rate": 4.9996222705552933e-05,
      "loss": 0.735,
      "step": 105
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.3488354682922363,
      "learning_rate": 4.9995850783649665e-05,
      "loss": 0.8344,
      "step": 110
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.1043323278427124,
      "learning_rate": 4.9995461402600593e-05,
      "loss": 0.8254,
      "step": 115
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.9382895827293396,
      "learning_rate": 4.9995054562677684e-05,
      "loss": 0.9179,
      "step": 120
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.2824612855911255,
      "learning_rate": 4.9994630264165107e-05,
      "loss": 0.8663,
      "step": 125
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.0491925477981567,
      "learning_rate": 4.999418850735923e-05,
      "loss": 0.9247,
      "step": 130
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.3642233610153198,
      "learning_rate": 4.99937292925686e-05,
      "loss": 0.8253,
      "step": 135
    },
    {
      "epoch": 0.02,
      "grad_norm": 3.747757911682129,
      "learning_rate": 4.9993252620113976e-05,
      "loss": 1.0245,
      "step": 140
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.299494981765747,
      "learning_rate": 4.999275849032832e-05,
      "loss": 0.8723,
      "step": 145
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.7195830345153809,
      "learning_rate": 4.999224690355675e-05,
      "loss": 1.0524,
      "step": 150
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.9922987222671509,
      "learning_rate": 4.9991717860156616e-05,
      "loss": 0.9502,
      "step": 155
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.0577458143234253,
      "learning_rate": 4.9991171360497437e-05,
      "loss": 1.0115,
      "step": 160
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.0001195669174194,
      "learning_rate": 4.999060740496093e-05,
      "loss": 1.1999,
      "step": 165
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.2456804513931274,
      "learning_rate": 4.999002599394102e-05,
      "loss": 0.8882,
      "step": 170
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.0445325374603271,
      "learning_rate": 4.9989427127843814e-05,
      "loss": 1.0615,
      "step": 175
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.2410887479782104,
      "learning_rate": 4.9988810807087584e-05,
      "loss": 1.1068,
      "step": 180
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.8935971260070801,
      "learning_rate": 4.998817703210285e-05,
      "loss": 0.6683,
      "step": 185
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.1614488363265991,
      "learning_rate": 4.9987525803332265e-05,
      "loss": 0.7446,
      "step": 190
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.9392004013061523,
      "learning_rate": 4.998685712123072e-05,
      "loss": 0.7397,
      "step": 195
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.0314444303512573,
      "learning_rate": 4.9986170986265266e-05,
      "loss": 1.3584,
      "step": 200
    },
    {
      "epoch": 0.03,
      "eval_loss": 0.8368077278137207,
      "eval_runtime": 96.5262,
      "eval_samples_per_second": 7.221,
      "eval_steps_per_second": 7.221,
      "step": 200
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.8964811563491821,
      "learning_rate": 4.998546739891516e-05,
      "loss": 0.9546,
      "step": 205
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.0679796934127808,
      "learning_rate": 4.998474635967185e-05,
      "loss": 0.864,
      "step": 210
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.2340985536575317,
      "learning_rate": 4.998400786903896e-05,
      "loss": 0.885,
      "step": 215
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.7219617366790771,
      "learning_rate": 4.9983251927532315e-05,
      "loss": 1.1069,
      "step": 220
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.1480705738067627,
      "learning_rate": 4.9982478535679924e-05,
      "loss": 1.0416,
      "step": 225
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.515589714050293,
      "learning_rate": 4.9981687694021996e-05,
      "loss": 1.1844,
      "step": 230
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.6687963008880615,
      "learning_rate": 4.998087940311091e-05,
      "loss": 0.8664,
      "step": 235
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.9256645441055298,
      "learning_rate": 4.998005366351125e-05,
      "loss": 1.0125,
      "step": 240
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.2500052452087402,
      "learning_rate": 4.997921047579978e-05,
      "loss": 1.1374,
      "step": 245
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.0543216466903687,
      "learning_rate": 4.9978349840565434e-05,
      "loss": 0.8502,
      "step": 250
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.3009012937545776,
      "learning_rate": 4.997747175840937e-05,
      "loss": 1.0357,
      "step": 255
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.8456661105155945,
      "learning_rate": 4.997657622994491e-05,
      "loss": 0.6883,
      "step": 260
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.5856515765190125,
      "learning_rate": 4.9975663255797555e-05,
      "loss": 0.7656,
      "step": 265
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.973818302154541,
      "learning_rate": 4.997473283660501e-05,
      "loss": 0.823,
      "step": 270
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.9960187673568726,
      "learning_rate": 4.997378497301715e-05,
      "loss": 0.8726,
      "step": 275
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.2900679111480713,
      "learning_rate": 4.997281966569604e-05,
      "loss": 0.9781,
      "step": 280
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.828894853591919,
      "learning_rate": 4.9971836915315926e-05,
      "loss": 0.8932,
      "step": 285
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.239621877670288,
      "learning_rate": 4.9970836722563256e-05,
      "loss": 1.2022,
      "step": 290
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.0117149353027344,
      "learning_rate": 4.996981908813664e-05,
      "loss": 0.8032,
      "step": 295
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.8861119747161865,
      "learning_rate": 4.996878401274687e-05,
      "loss": 1.0651,
      "step": 300
    },
    {
      "epoch": 0.05,
      "eval_loss": 0.8281473517417908,
      "eval_runtime": 96.5283,
      "eval_samples_per_second": 7.221,
      "eval_steps_per_second": 7.221,
      "step": 300
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.8583046197891235,
      "learning_rate": 4.996773149711693e-05,
      "loss": 0.8784,
      "step": 305
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.5717499256134033,
      "learning_rate": 4.9966661541981984e-05,
      "loss": 0.8395,
      "step": 310
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.982342004776001,
      "learning_rate": 4.9965574148089376e-05,
      "loss": 0.9869,
      "step": 315
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.9000777006149292,
      "learning_rate": 4.9964469316198633e-05,
      "loss": 0.8435,
      "step": 320
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.8733209371566772,
      "learning_rate": 4.9963347047081464e-05,
      "loss": 0.7281,
      "step": 325
    },
    {
      "epoch": 0.05,
      "grad_norm": 3.323739767074585,
      "learning_rate": 4.9962207341521746e-05,
      "loss": 1.1013,
      "step": 330
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.7102876901626587,
      "learning_rate": 4.996105020031554e-05,
      "loss": 0.8276,
      "step": 335
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.9196123480796814,
      "learning_rate": 4.995987562427109e-05,
      "loss": 0.8274,
      "step": 340
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.210099458694458,
      "learning_rate": 4.995868361420883e-05,
      "loss": 1.3257,
      "step": 345
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.8923581838607788,
      "learning_rate": 4.9957474170961335e-05,
      "loss": 0.6815,
      "step": 350
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.9576735496520996,
      "learning_rate": 4.9956247295373396e-05,
      "loss": 1.23,
      "step": 355
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.3774089813232422,
      "learning_rate": 4.995500298830196e-05,
      "loss": 1.0556,
      "step": 360
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.1523677110671997,
      "learning_rate": 4.995374125061614e-05,
      "loss": 1.1787,
      "step": 365
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.8310608863830566,
      "learning_rate": 4.9952462083197246e-05,
      "loss": 0.8525,
      "step": 370
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.9814196825027466,
      "learning_rate": 4.9951165486938765e-05,
      "loss": 0.8522,
      "step": 375
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.9878122210502625,
      "learning_rate": 4.994985146274633e-05,
      "loss": 0.6618,
      "step": 380
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.2652586698532104,
      "learning_rate": 4.994852001153777e-05,
      "loss": 1.0489,
      "step": 385
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.2940975427627563,
      "learning_rate": 4.994717113424307e-05,
      "loss": 1.104,
      "step": 390
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.9636249542236328,
      "learning_rate": 4.99458048318044e-05,
      "loss": 0.9228,
      "step": 395
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.8122813105583191,
      "learning_rate": 4.994442110517611e-05,
      "loss": 0.9209,
      "step": 400
    },
    {
      "epoch": 0.06,
      "eval_loss": 0.8184689879417419,
      "eval_runtime": 96.4572,
      "eval_samples_per_second": 7.226,
      "eval_steps_per_second": 7.226,
      "step": 400
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.8742052912712097,
      "learning_rate": 4.99430199553247e-05,
      "loss": 0.9608,
      "step": 405
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.5679522752761841,
      "learning_rate": 4.9941601383228835e-05,
      "loss": 0.5963,
      "step": 410
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.0234627723693848,
      "learning_rate": 4.994016538987938e-05,
      "loss": 0.8642,
      "step": 415
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.8581897616386414,
      "learning_rate": 4.993871197627934e-05,
      "loss": 0.8993,
      "step": 420
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.4666485786437988,
      "learning_rate": 4.9937241143443904e-05,
      "loss": 0.8565,
      "step": 425
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.1166578531265259,
      "learning_rate": 4.993575289240041e-05,
      "loss": 0.881,
      "step": 430
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.303992748260498,
      "learning_rate": 4.9934247224188393e-05,
      "loss": 0.9962,
      "step": 435
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.9011989235877991,
      "learning_rate": 4.993272413985952e-05,
      "loss": 0.9316,
      "step": 440
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.8321458101272583,
      "learning_rate": 4.993118364047764e-05,
      "loss": 0.7889,
      "step": 445
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.7780352234840393,
      "learning_rate": 4.992962572711877e-05,
      "loss": 0.8287,
      "step": 450
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.9090210199356079,
      "learning_rate": 4.992805040087108e-05,
      "loss": 0.7018,
      "step": 455
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.8694137334823608,
      "learning_rate": 4.9926457662834906e-05,
      "loss": 0.8484,
      "step": 460
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.6327371001243591,
      "learning_rate": 4.992484751412274e-05,
      "loss": 0.716,
      "step": 465
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.200668215751648,
      "learning_rate": 4.9923219955859254e-05,
      "loss": 0.9525,
      "step": 470
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.8530198931694031,
      "learning_rate": 4.9921574989181266e-05,
      "loss": 0.744,
      "step": 475
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.168479323387146,
      "learning_rate": 4.991991261523775e-05,
      "loss": 0.729,
      "step": 480
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.9499714970588684,
      "learning_rate": 4.9918232835189834e-05,
      "loss": 0.7725,
      "step": 485
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.8434467911720276,
      "learning_rate": 4.991653565021084e-05,
      "loss": 1.1558,
      "step": 490
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.7665804624557495,
      "learning_rate": 4.99148210614862e-05,
      "loss": 1.0208,
      "step": 495
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.5782546401023865,
      "learning_rate": 4.991308907021353e-05,
      "loss": 0.8306,
      "step": 500
    },
    {
      "epoch": 0.08,
      "eval_loss": 0.8132078051567078,
      "eval_runtime": 96.433,
      "eval_samples_per_second": 7.228,
      "eval_steps_per_second": 7.228,
      "step": 500
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.0821778774261475,
      "learning_rate": 4.9911339677602584e-05,
      "loss": 0.9503,
      "step": 505
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.5409029126167297,
      "learning_rate": 4.99095728848753e-05,
      "loss": 0.8586,
      "step": 510
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.9011789560317993,
      "learning_rate": 4.990778869326575e-05,
      "loss": 0.7981,
      "step": 515
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.0092263221740723,
      "learning_rate": 4.990598710402013e-05,
      "loss": 1.0174,
      "step": 520
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.4362307786941528,
      "learning_rate": 4.9904168118396844e-05,
      "loss": 0.8373,
      "step": 525
    },
    {
      "epoch": 0.08,
      "grad_norm": 2.1772639751434326,
      "learning_rate": 4.9902331737666414e-05,
      "loss": 0.9599,
      "step": 530
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.9610542058944702,
      "learning_rate": 4.990047796311151e-05,
      "loss": 0.6895,
      "step": 535
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.9922348260879517,
      "learning_rate": 4.989860679602698e-05,
      "loss": 0.7315,
      "step": 540
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.2409151792526245,
      "learning_rate": 4.9896718237719785e-05,
      "loss": 0.8574,
      "step": 545
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.016333818435669,
      "learning_rate": 4.9894812289509046e-05,
      "loss": 1.1248,
      "step": 550
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.9131489396095276,
      "learning_rate": 4.989288895272604e-05,
      "loss": 0.9847,
      "step": 555
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.215469479560852,
      "learning_rate": 4.989094822871419e-05,
      "loss": 0.912,
      "step": 560
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.0536105632781982,
      "learning_rate": 4.988899011882903e-05,
      "loss": 0.8425,
      "step": 565
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.9705311059951782,
      "learning_rate": 4.988701462443829e-05,
      "loss": 0.9385,
      "step": 570
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.2488442659378052,
      "learning_rate": 4.98850217469218e-05,
      "loss": 0.7865,
      "step": 575
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.7318600416183472,
      "learning_rate": 4.988301148767157e-05,
      "loss": 0.8231,
      "step": 580
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.8247858881950378,
      "learning_rate": 4.9880983848091704e-05,
      "loss": 0.8553,
      "step": 585
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.858172595500946,
      "learning_rate": 4.987893882959849e-05,
      "loss": 1.3952,
      "step": 590
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.2286418676376343,
      "learning_rate": 4.987687643362033e-05,
      "loss": 0.837,
      "step": 595
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.034350872039795,
      "learning_rate": 4.9874796661597765e-05,
      "loss": 0.9175,
      "step": 600
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.8063747882843018,
      "eval_runtime": 96.4224,
      "eval_samples_per_second": 7.229,
      "eval_steps_per_second": 7.229,
      "step": 600
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.7192366123199463,
      "learning_rate": 4.987269951498348e-05,
      "loss": 0.8563,
      "step": 605
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.2645854949951172,
      "learning_rate": 4.98705849952423e-05,
      "loss": 0.6663,
      "step": 610
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.0610381364822388,
      "learning_rate": 4.9868453103851176e-05,
      "loss": 0.8452,
      "step": 615
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.8550002574920654,
      "learning_rate": 4.986630384229919e-05,
      "loss": 0.8894,
      "step": 620
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.7490519285202026,
      "learning_rate": 4.986413721208757e-05,
      "loss": 0.9106,
      "step": 625
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.557860255241394,
      "learning_rate": 4.986195321472965e-05,
      "loss": 0.685,
      "step": 630
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.7450752258300781,
      "learning_rate": 4.9859751851750934e-05,
      "loss": 0.8472,
      "step": 635
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.176376461982727,
      "learning_rate": 4.985753312468903e-05,
      "loss": 1.0197,
      "step": 640
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.0625300407409668,
      "learning_rate": 4.985529703509367e-05,
      "loss": 0.9685,
      "step": 645
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.8808372616767883,
      "learning_rate": 4.985304358452672e-05,
      "loss": 0.8612,
      "step": 650
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.8110201954841614,
      "learning_rate": 4.985077277456218e-05,
      "loss": 0.8401,
      "step": 655
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.9364888072013855,
      "learning_rate": 4.984848460678618e-05,
      "loss": 0.6197,
      "step": 660
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.0113518238067627,
      "learning_rate": 4.984617908279694e-05,
      "loss": 0.9889,
      "step": 665
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.1148868799209595,
      "learning_rate": 4.984385620420485e-05,
      "loss": 0.9558,
      "step": 670
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.9506175518035889,
      "learning_rate": 4.984151597263238e-05,
      "loss": 0.7323,
      "step": 675
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.0044193267822266,
      "learning_rate": 4.983915838971415e-05,
      "loss": 0.7504,
      "step": 680
    },
    {
      "epoch": 0.11,
      "grad_norm": 2.2674214839935303,
      "learning_rate": 4.9836783457096875e-05,
      "loss": 1.032,
      "step": 685
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.4945333003997803,
      "learning_rate": 4.983439117643942e-05,
      "loss": 1.0359,
      "step": 690
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.9860715866088867,
      "learning_rate": 4.9831981549412744e-05,
      "loss": 1.1152,
      "step": 695
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.8287227153778076,
      "learning_rate": 4.982955457769992e-05,
      "loss": 0.8157,
      "step": 700
    },
    {
      "epoch": 0.11,
      "eval_loss": 0.8022791743278503,
      "eval_runtime": 96.5324,
      "eval_samples_per_second": 7.22,
      "eval_steps_per_second": 7.22,
      "step": 700
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.9216273427009583,
      "learning_rate": 4.9827110262996144e-05,
      "loss": 0.8395,
      "step": 705
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.7642357349395752,
      "learning_rate": 4.982464860700874e-05,
      "loss": 0.8817,
      "step": 710
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.8851175308227539,
      "learning_rate": 4.982216961145711e-05,
      "loss": 0.8558,
      "step": 715
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.44226109981536865,
      "learning_rate": 4.98196732780728e-05,
      "loss": 0.882,
      "step": 720
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.8005027174949646,
      "learning_rate": 4.981715960859945e-05,
      "loss": 0.8835,
      "step": 725
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.7451304793357849,
      "learning_rate": 4.981462860479281e-05,
      "loss": 0.8551,
      "step": 730
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.1069347858428955,
      "learning_rate": 4.9812080268420745e-05,
      "loss": 0.999,
      "step": 735
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.8892244100570679,
      "learning_rate": 4.980951460126322e-05,
      "loss": 1.012,
      "step": 740
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.8935977816581726,
      "learning_rate": 4.9806931605112305e-05,
      "loss": 0.9911,
      "step": 745
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.8456961512565613,
      "learning_rate": 4.9804331281772176e-05,
      "loss": 0.7595,
      "step": 750
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.78443443775177,
      "learning_rate": 4.980171363305911e-05,
      "loss": 0.8308,
      "step": 755
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.0028038024902344,
      "learning_rate": 4.979907866080149e-05,
      "loss": 0.9637,
      "step": 760
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.1801577806472778,
      "learning_rate": 4.9796426366839786e-05,
      "loss": 0.6159,
      "step": 765
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.8370681405067444,
      "learning_rate": 4.979375675302659e-05,
      "loss": 0.9276,
      "step": 770
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.8605382442474365,
      "learning_rate": 4.979106982122658e-05,
      "loss": 1.1077,
      "step": 775
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.7788259387016296,
      "learning_rate": 4.978836557331652e-05,
      "loss": 0.8172,
      "step": 780
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.4312686920166016,
      "learning_rate": 4.978564401118528e-05,
      "loss": 0.8759,
      "step": 785
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.9109662175178528,
      "learning_rate": 4.978290513673381e-05,
      "loss": 0.947,
      "step": 790
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.1819065809249878,
      "learning_rate": 4.9780148951875195e-05,
      "loss": 0.7364,
      "step": 795
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.9400575160980225,
      "learning_rate": 4.977737545853455e-05,
      "loss": 0.9469,
      "step": 800
    },
    {
      "epoch": 0.13,
      "eval_loss": 0.7995806932449341,
      "eval_runtime": 96.5877,
      "eval_samples_per_second": 7.216,
      "eval_steps_per_second": 7.216,
      "step": 800
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.693812370300293,
      "learning_rate": 4.9774584658649126e-05,
      "loss": 0.9433,
      "step": 805
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.0892895460128784,
      "learning_rate": 4.9771776554168234e-05,
      "loss": 0.7027,
      "step": 810
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.9118362665176392,
      "learning_rate": 4.976895114705329e-05,
      "loss": 0.9468,
      "step": 815
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.8032681345939636,
      "learning_rate": 4.976610843927779e-05,
      "loss": 0.7927,
      "step": 820
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.168225646018982,
      "learning_rate": 4.976324843282732e-05,
      "loss": 0.9673,
      "step": 825
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.077602744102478,
      "learning_rate": 4.976037112969953e-05,
      "loss": 0.9156,
      "step": 830
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.8643108606338501,
      "learning_rate": 4.9757476531904165e-05,
      "loss": 0.6999,
      "step": 835
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.933397650718689,
      "learning_rate": 4.975456464146306e-05,
      "loss": 0.8828,
      "step": 840
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.7036295533180237,
      "learning_rate": 4.975163546041011e-05,
      "loss": 0.8709,
      "step": 845
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5974694490432739,
      "learning_rate": 4.974868899079128e-05,
      "loss": 0.7594,
      "step": 850
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.7244943380355835,
      "learning_rate": 4.974572523466465e-05,
      "loss": 0.8714,
      "step": 855
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5783522725105286,
      "learning_rate": 4.9742744194100345e-05,
      "loss": 0.8941,
      "step": 860
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.7480617761611938,
      "learning_rate": 4.973974587118055e-05,
      "loss": 0.9798,
      "step": 865
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.7548874020576477,
      "learning_rate": 4.973673026799956e-05,
      "loss": 0.7767,
      "step": 870
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.7075071930885315,
      "learning_rate": 4.97336973866637e-05,
      "loss": 0.7779,
      "step": 875
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.7042987942695618,
      "learning_rate": 4.97306472292914e-05,
      "loss": 0.8249,
      "step": 880
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.0242459774017334,
      "learning_rate": 4.972757979801313e-05,
      "loss": 0.9223,
      "step": 885
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.6138095259666443,
      "learning_rate": 4.9724495094971436e-05,
      "loss": 0.9842,
      "step": 890
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.7905042767524719,
      "learning_rate": 4.9721393122320925e-05,
      "loss": 0.8738,
      "step": 895
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.9658048748970032,
      "learning_rate": 4.9718273882228265e-05,
      "loss": 0.8872,
      "step": 900
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.7954564690589905,
      "eval_runtime": 96.643,
      "eval_samples_per_second": 7.212,
      "eval_steps_per_second": 7.212,
      "step": 900
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.8425014019012451,
      "learning_rate": 4.97151373768722e-05,
      "loss": 0.778,
      "step": 905
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.5527231693267822,
      "learning_rate": 4.971198360844351e-05,
      "loss": 0.8332,
      "step": 910
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.7870334386825562,
      "learning_rate": 4.9708812579145056e-05,
      "loss": 0.9265,
      "step": 915
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.9935321807861328,
      "learning_rate": 4.970562429119173e-05,
      "loss": 0.7243,
      "step": 920
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.9546892046928406,
      "learning_rate": 4.970241874681051e-05,
      "loss": 0.9908,
      "step": 925
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.7340118885040283,
      "learning_rate": 4.969919594824039e-05,
      "loss": 0.7932,
      "step": 930
    },
    {
      "epoch": 0.15,
      "grad_norm": 5.1686015129089355,
      "learning_rate": 4.9695955897732453e-05,
      "loss": 0.9842,
      "step": 935
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.9721456170082092,
      "learning_rate": 4.9692698597549815e-05,
      "loss": 0.9271,
      "step": 940
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.6477334499359131,
      "learning_rate": 4.9689424049967623e-05,
      "loss": 0.934,
      "step": 945
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.0759055614471436,
      "learning_rate": 4.968613225727311e-05,
      "loss": 1.0465,
      "step": 950
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.7222158908843994,
      "learning_rate": 4.968282322176552e-05,
      "loss": 0.7732,
      "step": 955
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.8591343760490417,
      "learning_rate": 4.9679496945756155e-05,
      "loss": 0.9062,
      "step": 960
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.8495111465454102,
      "learning_rate": 4.967615343156837e-05,
      "loss": 0.8861,
      "step": 965
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.6847331523895264,
      "learning_rate": 4.967279268153753e-05,
      "loss": 0.8001,
      "step": 970
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.690113365650177,
      "learning_rate": 4.9669414698011074e-05,
      "loss": 0.7378,
      "step": 975
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.8349626064300537,
      "learning_rate": 4.9666019483348456e-05,
      "loss": 0.7193,
      "step": 980
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6444108486175537,
      "learning_rate": 4.966260703992116e-05,
      "loss": 0.8729,
      "step": 985
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.9515655040740967,
      "learning_rate": 4.965917737011274e-05,
      "loss": 0.7532,
      "step": 990
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.8138986229896545,
      "learning_rate": 4.965573047631873e-05,
      "loss": 1.0124,
      "step": 995
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.0182080268859863,
      "learning_rate": 4.9652266360946745e-05,
      "loss": 0.8842,
      "step": 1000
    },
    {
      "epoch": 0.16,
      "eval_loss": 0.7912728190422058,
      "eval_runtime": 96.5004,
      "eval_samples_per_second": 7.223,
      "eval_steps_per_second": 7.223,
      "step": 1000
    }
  ],
  "logging_steps": 5,
  "max_steps": 18795,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 4.3155317587968e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}