zephyr-7b-gemma-sft-5p-2048 / trainer_state.json
Qin Liu
Model save
8f6eefa verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.997697620874904,
"eval_steps": 500,
"global_step": 1953,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"grad_norm": 72.47212998792584,
"learning_rate": 1.020408163265306e-06,
"loss": 17.0139,
"step": 1
},
{
"epoch": 0.01,
"grad_norm": 72.19709441394318,
"learning_rate": 5.102040816326531e-06,
"loss": 16.8099,
"step": 5
},
{
"epoch": 0.02,
"grad_norm": 74.11072818499085,
"learning_rate": 1.0204081632653061e-05,
"loss": 16.7768,
"step": 10
},
{
"epoch": 0.02,
"grad_norm": 67.88747859998395,
"learning_rate": 1.5306122448979594e-05,
"loss": 15.8745,
"step": 15
},
{
"epoch": 0.03,
"grad_norm": 57.15585576506464,
"learning_rate": 2.0408163265306123e-05,
"loss": 13.4197,
"step": 20
},
{
"epoch": 0.04,
"grad_norm": 52.16705016348157,
"learning_rate": 2.5510204081632654e-05,
"loss": 10.8782,
"step": 25
},
{
"epoch": 0.05,
"grad_norm": 34.872891212782,
"learning_rate": 3.061224489795919e-05,
"loss": 8.0236,
"step": 30
},
{
"epoch": 0.05,
"grad_norm": 20.96608544121856,
"learning_rate": 3.571428571428572e-05,
"loss": 5.7602,
"step": 35
},
{
"epoch": 0.06,
"grad_norm": 14.39165011334111,
"learning_rate": 4.0816326530612245e-05,
"loss": 4.4129,
"step": 40
},
{
"epoch": 0.07,
"grad_norm": 9.537119159153542,
"learning_rate": 4.591836734693878e-05,
"loss": 3.4241,
"step": 45
},
{
"epoch": 0.08,
"grad_norm": 10.051565364855378,
"learning_rate": 5.102040816326531e-05,
"loss": 2.7711,
"step": 50
},
{
"epoch": 0.08,
"grad_norm": 5.902644991920862,
"learning_rate": 5.6122448979591836e-05,
"loss": 2.1132,
"step": 55
},
{
"epoch": 0.09,
"grad_norm": 5.282533867200368,
"learning_rate": 6.122448979591838e-05,
"loss": 1.7932,
"step": 60
},
{
"epoch": 0.1,
"grad_norm": 2.849440118853736,
"learning_rate": 6.63265306122449e-05,
"loss": 1.5357,
"step": 65
},
{
"epoch": 0.11,
"grad_norm": 6.395981446361856,
"learning_rate": 7.142857142857143e-05,
"loss": 1.4488,
"step": 70
},
{
"epoch": 0.12,
"grad_norm": 1.9290320550828686,
"learning_rate": 7.653061224489796e-05,
"loss": 1.3654,
"step": 75
},
{
"epoch": 0.12,
"grad_norm": 1.9509560155111505,
"learning_rate": 8.163265306122449e-05,
"loss": 1.2287,
"step": 80
},
{
"epoch": 0.13,
"grad_norm": 1.9800497930938712,
"learning_rate": 8.673469387755102e-05,
"loss": 1.2863,
"step": 85
},
{
"epoch": 0.14,
"grad_norm": 1.7120992753385207,
"learning_rate": 9.183673469387756e-05,
"loss": 1.1712,
"step": 90
},
{
"epoch": 0.15,
"grad_norm": 2.2653861656665453,
"learning_rate": 9.693877551020408e-05,
"loss": 1.2174,
"step": 95
},
{
"epoch": 0.15,
"grad_norm": 3.453167617907916,
"learning_rate": 0.00010204081632653062,
"loss": 1.1714,
"step": 100
},
{
"epoch": 0.16,
"grad_norm": 1.547620666252296,
"learning_rate": 0.00010714285714285715,
"loss": 1.1168,
"step": 105
},
{
"epoch": 0.17,
"grad_norm": 1.6713272237258214,
"learning_rate": 0.00011224489795918367,
"loss": 1.0868,
"step": 110
},
{
"epoch": 0.18,
"grad_norm": 1.5402309108002492,
"learning_rate": 0.00011734693877551022,
"loss": 1.1351,
"step": 115
},
{
"epoch": 0.18,
"grad_norm": 1.5970274628335959,
"learning_rate": 0.00012244897959183676,
"loss": 1.0946,
"step": 120
},
{
"epoch": 0.19,
"grad_norm": 1.6342350847388514,
"learning_rate": 0.00012755102040816328,
"loss": 1.1211,
"step": 125
},
{
"epoch": 0.2,
"grad_norm": 1.6632080501970268,
"learning_rate": 0.0001326530612244898,
"loss": 1.0501,
"step": 130
},
{
"epoch": 0.21,
"grad_norm": 1.4060120288881937,
"learning_rate": 0.00013775510204081635,
"loss": 1.0956,
"step": 135
},
{
"epoch": 0.21,
"grad_norm": 1.801531788644497,
"learning_rate": 0.00014285714285714287,
"loss": 1.072,
"step": 140
},
{
"epoch": 0.22,
"grad_norm": 1.520734386032582,
"learning_rate": 0.0001479591836734694,
"loss": 1.1119,
"step": 145
},
{
"epoch": 0.23,
"grad_norm": 1.840843283748014,
"learning_rate": 0.0001530612244897959,
"loss": 0.9969,
"step": 150
},
{
"epoch": 0.24,
"grad_norm": 2.07792070441396,
"learning_rate": 0.00015816326530612246,
"loss": 1.0208,
"step": 155
},
{
"epoch": 0.25,
"grad_norm": 1.9723613912374163,
"learning_rate": 0.00016326530612244898,
"loss": 0.9975,
"step": 160
},
{
"epoch": 0.25,
"grad_norm": 2.8108507027898844,
"learning_rate": 0.00016836734693877553,
"loss": 1.0388,
"step": 165
},
{
"epoch": 0.26,
"grad_norm": 2.049204774598262,
"learning_rate": 0.00017346938775510205,
"loss": 0.9837,
"step": 170
},
{
"epoch": 0.27,
"grad_norm": 2.013771271987093,
"learning_rate": 0.0001785714285714286,
"loss": 0.956,
"step": 175
},
{
"epoch": 0.28,
"grad_norm": 1.6734633410801107,
"learning_rate": 0.00018367346938775512,
"loss": 1.0388,
"step": 180
},
{
"epoch": 0.28,
"grad_norm": 2.3629530195645043,
"learning_rate": 0.00018877551020408164,
"loss": 1.0407,
"step": 185
},
{
"epoch": 0.29,
"grad_norm": 2.0061060540694267,
"learning_rate": 0.00019387755102040816,
"loss": 1.0528,
"step": 190
},
{
"epoch": 0.3,
"grad_norm": 1.8957278214013087,
"learning_rate": 0.0001989795918367347,
"loss": 1.0637,
"step": 195
},
{
"epoch": 0.31,
"grad_norm": 1.659431161796715,
"learning_rate": 0.00019999744233089168,
"loss": 1.0197,
"step": 200
},
{
"epoch": 0.31,
"grad_norm": 1.9623902009306882,
"learning_rate": 0.00019998705202436978,
"loss": 1.0171,
"step": 205
},
{
"epoch": 0.32,
"grad_norm": 1.666768664331948,
"learning_rate": 0.0001999686700559419,
"loss": 0.9411,
"step": 210
},
{
"epoch": 0.33,
"grad_norm": 1.6263956387135465,
"learning_rate": 0.00019994229789482308,
"loss": 1.0452,
"step": 215
},
{
"epoch": 0.34,
"grad_norm": 2.104534739444239,
"learning_rate": 0.00019990793764886012,
"loss": 1.0318,
"step": 220
},
{
"epoch": 0.35,
"grad_norm": 1.5223676518998526,
"learning_rate": 0.0001998655920643634,
"loss": 0.9393,
"step": 225
},
{
"epoch": 0.35,
"grad_norm": 1.399019073099987,
"learning_rate": 0.000199815264525887,
"loss": 0.97,
"step": 230
},
{
"epoch": 0.36,
"grad_norm": 1.8651896389488705,
"learning_rate": 0.00019975695905595855,
"loss": 1.0187,
"step": 235
},
{
"epoch": 0.37,
"grad_norm": 1.536205961219932,
"learning_rate": 0.00019969068031475744,
"loss": 0.9716,
"step": 240
},
{
"epoch": 0.38,
"grad_norm": 1.7078218228495357,
"learning_rate": 0.0001996164335997425,
"loss": 0.9959,
"step": 245
},
{
"epoch": 0.38,
"grad_norm": 1.3467510953394168,
"learning_rate": 0.0001995342248452285,
"loss": 0.9602,
"step": 250
},
{
"epoch": 0.39,
"grad_norm": 1.9084544882836239,
"learning_rate": 0.00019944406062191204,
"loss": 0.9775,
"step": 255
},
{
"epoch": 0.4,
"grad_norm": 1.5574610066347279,
"learning_rate": 0.000199345948136346,
"loss": 1.0177,
"step": 260
},
{
"epoch": 0.41,
"grad_norm": 2.109450536237562,
"learning_rate": 0.00019923989523036394,
"loss": 0.9819,
"step": 265
},
{
"epoch": 0.41,
"grad_norm": 2.7828828000504573,
"learning_rate": 0.00019912591038045307,
"loss": 0.9707,
"step": 270
},
{
"epoch": 0.42,
"grad_norm": 1.3138848135299093,
"learning_rate": 0.0001990040026970768,
"loss": 0.9909,
"step": 275
},
{
"epoch": 0.43,
"grad_norm": 1.5708872027230352,
"learning_rate": 0.0001988741819239467,
"loss": 0.9744,
"step": 280
},
{
"epoch": 0.44,
"grad_norm": 1.7915877912980807,
"learning_rate": 0.0001987364584372435,
"loss": 0.9629,
"step": 285
},
{
"epoch": 0.45,
"grad_norm": 1.7688040681479107,
"learning_rate": 0.00019859084324478791,
"loss": 1.0066,
"step": 290
},
{
"epoch": 0.45,
"grad_norm": 1.2087916028360217,
"learning_rate": 0.00019843734798516077,
"loss": 0.9273,
"step": 295
},
{
"epoch": 0.46,
"grad_norm": 1.7217374396550518,
"learning_rate": 0.00019827598492677283,
"loss": 0.9599,
"step": 300
},
{
"epoch": 0.47,
"grad_norm": 1.1827063296813014,
"learning_rate": 0.000198106766966884,
"loss": 0.9526,
"step": 305
},
{
"epoch": 0.48,
"grad_norm": 1.275288932081838,
"learning_rate": 0.0001979297076305728,
"loss": 0.9351,
"step": 310
},
{
"epoch": 0.48,
"grad_norm": 1.4572732094413907,
"learning_rate": 0.00019774482106965513,
"loss": 0.9916,
"step": 315
},
{
"epoch": 0.49,
"grad_norm": 1.3548000574556105,
"learning_rate": 0.00019755212206155318,
"loss": 0.9894,
"step": 320
},
{
"epoch": 0.5,
"grad_norm": 1.4221295720770253,
"learning_rate": 0.00019735162600811447,
"loss": 0.9147,
"step": 325
},
{
"epoch": 0.51,
"grad_norm": 1.1147927135270754,
"learning_rate": 0.00019714334893438062,
"loss": 0.9173,
"step": 330
},
{
"epoch": 0.51,
"grad_norm": 1.5740508551311372,
"learning_rate": 0.00019692730748730662,
"loss": 1.0049,
"step": 335
},
{
"epoch": 0.52,
"grad_norm": 1.4361894716708776,
"learning_rate": 0.0001967035189344303,
"loss": 0.9772,
"step": 340
},
{
"epoch": 0.53,
"grad_norm": 1.5979099481664976,
"learning_rate": 0.00019647200116249214,
"loss": 0.9734,
"step": 345
},
{
"epoch": 0.54,
"grad_norm": 1.1416579629890438,
"learning_rate": 0.00019623277267600574,
"loss": 0.9695,
"step": 350
},
{
"epoch": 0.54,
"grad_norm": 1.2859502643647958,
"learning_rate": 0.0001959858525957786,
"loss": 0.9726,
"step": 355
},
{
"epoch": 0.55,
"grad_norm": 1.654528585364595,
"learning_rate": 0.00019573126065738415,
"loss": 0.9099,
"step": 360
},
{
"epoch": 0.56,
"grad_norm": 1.6432235938460993,
"learning_rate": 0.00019546901720958405,
"loss": 0.993,
"step": 365
},
{
"epoch": 0.57,
"grad_norm": 1.184408195362666,
"learning_rate": 0.00019519914321270196,
"loss": 0.983,
"step": 370
},
{
"epoch": 0.58,
"grad_norm": 1.364894780646226,
"learning_rate": 0.00019492166023694823,
"loss": 0.9385,
"step": 375
},
{
"epoch": 0.58,
"grad_norm": 1.2140596195120288,
"learning_rate": 0.0001946365904606957,
"loss": 0.928,
"step": 380
},
{
"epoch": 0.59,
"grad_norm": 1.2023189029574775,
"learning_rate": 0.00019434395666870734,
"loss": 0.9497,
"step": 385
},
{
"epoch": 0.6,
"grad_norm": 1.5293781545066814,
"learning_rate": 0.00019404378225031482,
"loss": 0.9845,
"step": 390
},
{
"epoch": 0.61,
"grad_norm": 1.2719601216605032,
"learning_rate": 0.00019373609119754926,
"loss": 0.9535,
"step": 395
},
{
"epoch": 0.61,
"grad_norm": 1.1767287730617413,
"learning_rate": 0.00019342090810322361,
"loss": 0.9669,
"step": 400
},
{
"epoch": 0.62,
"grad_norm": 1.5092663681250142,
"learning_rate": 0.00019309825815896697,
"loss": 0.9097,
"step": 405
},
{
"epoch": 0.63,
"grad_norm": 1.065331569108817,
"learning_rate": 0.00019276816715321107,
"loss": 0.9257,
"step": 410
},
{
"epoch": 0.64,
"grad_norm": 1.0551712975999572,
"learning_rate": 0.00019243066146912914,
"loss": 0.9179,
"step": 415
},
{
"epoch": 0.64,
"grad_norm": 1.280851911643696,
"learning_rate": 0.00019208576808252726,
"loss": 0.9322,
"step": 420
},
{
"epoch": 0.65,
"grad_norm": 1.1957795064970969,
"learning_rate": 0.00019173351455968805,
"loss": 0.9472,
"step": 425
},
{
"epoch": 0.66,
"grad_norm": 1.6691262118179164,
"learning_rate": 0.00019137392905516757,
"loss": 0.9833,
"step": 430
},
{
"epoch": 0.67,
"grad_norm": 1.2087456453830114,
"learning_rate": 0.0001910070403095449,
"loss": 0.9554,
"step": 435
},
{
"epoch": 0.68,
"grad_norm": 1.2175939573659607,
"learning_rate": 0.00019063287764712513,
"loss": 0.9844,
"step": 440
},
{
"epoch": 0.68,
"grad_norm": 1.4498433195544413,
"learning_rate": 0.00019025147097359528,
"loss": 0.9467,
"step": 445
},
{
"epoch": 0.69,
"grad_norm": 1.429864892305334,
"learning_rate": 0.00018986285077363446,
"loss": 0.9309,
"step": 450
},
{
"epoch": 0.7,
"grad_norm": 1.170651014792718,
"learning_rate": 0.00018946704810847689,
"loss": 0.9234,
"step": 455
},
{
"epoch": 0.71,
"grad_norm": 1.3282215747364223,
"learning_rate": 0.00018906409461342952,
"loss": 0.9536,
"step": 460
},
{
"epoch": 0.71,
"grad_norm": 1.1845632567557303,
"learning_rate": 0.00018865402249534347,
"loss": 0.9772,
"step": 465
},
{
"epoch": 0.72,
"grad_norm": 1.2127271540342577,
"learning_rate": 0.00018823686453003973,
"loss": 0.9523,
"step": 470
},
{
"epoch": 0.73,
"grad_norm": 1.2360135960506637,
"learning_rate": 0.00018781265405968972,
"loss": 0.9135,
"step": 475
},
{
"epoch": 0.74,
"grad_norm": 1.5041667778892749,
"learning_rate": 0.0001873814249901501,
"loss": 0.9625,
"step": 480
},
{
"epoch": 0.74,
"grad_norm": 1.1504429942546481,
"learning_rate": 0.00018694321178825286,
"loss": 0.9363,
"step": 485
},
{
"epoch": 0.75,
"grad_norm": 1.0319491329942774,
"learning_rate": 0.00018649804947905055,
"loss": 0.9054,
"step": 490
},
{
"epoch": 0.76,
"grad_norm": 1.2204274089754297,
"learning_rate": 0.0001860459736430169,
"loss": 0.9635,
"step": 495
},
{
"epoch": 0.77,
"grad_norm": 1.0611368485238843,
"learning_rate": 0.00018558702041320273,
"loss": 0.9445,
"step": 500
},
{
"epoch": 0.78,
"grad_norm": 1.1018739389426906,
"learning_rate": 0.00018512122647234812,
"loss": 0.9289,
"step": 505
},
{
"epoch": 0.78,
"grad_norm": 0.9882389573927103,
"learning_rate": 0.0001846486290499505,
"loss": 0.9911,
"step": 510
},
{
"epoch": 0.79,
"grad_norm": 1.2994396811323112,
"learning_rate": 0.0001841692659192889,
"loss": 0.9264,
"step": 515
},
{
"epoch": 0.8,
"grad_norm": 1.5236972647383147,
"learning_rate": 0.00018368317539440492,
"loss": 0.9563,
"step": 520
},
{
"epoch": 0.81,
"grad_norm": 1.1383444262695794,
"learning_rate": 0.0001831903963270404,
"loss": 0.977,
"step": 525
},
{
"epoch": 0.81,
"grad_norm": 1.110789605109835,
"learning_rate": 0.00018269096810353205,
"loss": 0.9388,
"step": 530
},
{
"epoch": 0.82,
"grad_norm": 1.6690032269651018,
"learning_rate": 0.00018218493064166353,
"loss": 0.923,
"step": 535
},
{
"epoch": 0.83,
"grad_norm": 1.2741603156202803,
"learning_rate": 0.00018167232438747485,
"loss": 0.959,
"step": 540
},
{
"epoch": 0.84,
"grad_norm": 1.1359315426783947,
"learning_rate": 0.00018115319031202965,
"loss": 0.958,
"step": 545
},
{
"epoch": 0.84,
"grad_norm": 1.1462965424102658,
"learning_rate": 0.00018062756990814058,
"loss": 0.9206,
"step": 550
},
{
"epoch": 0.85,
"grad_norm": 0.9795818536509971,
"learning_rate": 0.00018009550518705285,
"loss": 0.9027,
"step": 555
},
{
"epoch": 0.86,
"grad_norm": 0.9867688236460216,
"learning_rate": 0.00017955703867508633,
"loss": 0.9283,
"step": 560
},
{
"epoch": 0.87,
"grad_norm": 1.0077012911846719,
"learning_rate": 0.00017901221341023673,
"loss": 0.9516,
"step": 565
},
{
"epoch": 0.87,
"grad_norm": 1.3421028406940887,
"learning_rate": 0.00017846107293873555,
"loss": 0.9121,
"step": 570
},
{
"epoch": 0.88,
"grad_norm": 1.4547843706488663,
"learning_rate": 0.0001779036613115696,
"loss": 0.8875,
"step": 575
},
{
"epoch": 0.89,
"grad_norm": 1.1060535265201301,
"learning_rate": 0.00017734002308096014,
"loss": 0.9554,
"step": 580
},
{
"epoch": 0.9,
"grad_norm": 1.5055847688809278,
"learning_rate": 0.00017677020329680203,
"loss": 0.9173,
"step": 585
},
{
"epoch": 0.91,
"grad_norm": 1.1611583032421355,
"learning_rate": 0.00017619424750306287,
"loss": 0.9086,
"step": 590
},
{
"epoch": 0.91,
"grad_norm": 1.1353445717142476,
"learning_rate": 0.00017561220173414297,
"loss": 0.967,
"step": 595
},
{
"epoch": 0.92,
"grad_norm": 0.9563792857101335,
"learning_rate": 0.00017502411251119586,
"loss": 0.9155,
"step": 600
},
{
"epoch": 0.93,
"grad_norm": 1.1337314999879342,
"learning_rate": 0.00017443002683841002,
"loss": 0.8905,
"step": 605
},
{
"epoch": 0.94,
"grad_norm": 1.0287944149688226,
"learning_rate": 0.00017382999219925203,
"loss": 0.9092,
"step": 610
},
{
"epoch": 0.94,
"grad_norm": 1.2022884801513027,
"learning_rate": 0.00017322405655267122,
"loss": 0.8703,
"step": 615
},
{
"epoch": 0.95,
"grad_norm": 1.2018380484271611,
"learning_rate": 0.0001726122683292667,
"loss": 0.9769,
"step": 620
},
{
"epoch": 0.96,
"grad_norm": 1.2278521149338097,
"learning_rate": 0.0001719946764274162,
"loss": 0.9632,
"step": 625
},
{
"epoch": 0.97,
"grad_norm": 1.132615620517034,
"learning_rate": 0.00017137133020936782,
"loss": 0.943,
"step": 630
},
{
"epoch": 0.97,
"grad_norm": 1.0891584621905475,
"learning_rate": 0.00017074227949729481,
"loss": 0.9249,
"step": 635
},
{
"epoch": 0.98,
"grad_norm": 1.1417202547948113,
"learning_rate": 0.00017010757456931334,
"loss": 0.9055,
"step": 640
},
{
"epoch": 0.99,
"grad_norm": 0.9346881604970835,
"learning_rate": 0.0001694672661554638,
"loss": 0.9336,
"step": 645
},
{
"epoch": 1.0,
"grad_norm": 1.1301994877879067,
"learning_rate": 0.0001688214054336563,
"loss": 0.9062,
"step": 650
},
{
"epoch": 1.0,
"eval_loss": 1.2441989183425903,
"eval_runtime": 252.9523,
"eval_samples_per_second": 9.132,
"eval_steps_per_second": 0.573,
"step": 651
},
{
"epoch": 1.01,
"grad_norm": 1.179898601273253,
"learning_rate": 0.00016817004402558012,
"loss": 0.9027,
"step": 655
},
{
"epoch": 1.01,
"grad_norm": 0.9417315541112381,
"learning_rate": 0.0001675132339925776,
"loss": 0.9119,
"step": 660
},
{
"epoch": 1.02,
"grad_norm": 1.0837703721555252,
"learning_rate": 0.0001668510278314833,
"loss": 0.86,
"step": 665
},
{
"epoch": 1.03,
"grad_norm": 1.1272174437541118,
"learning_rate": 0.00016618347847042778,
"loss": 0.8711,
"step": 670
},
{
"epoch": 1.04,
"grad_norm": 1.1359544360875102,
"learning_rate": 0.00016551063926460748,
"loss": 0.8776,
"step": 675
},
{
"epoch": 1.04,
"grad_norm": 1.1917117624970714,
"learning_rate": 0.00016483256399202006,
"loss": 0.9209,
"step": 680
},
{
"epoch": 1.05,
"grad_norm": 1.0249074428213159,
"learning_rate": 0.00016414930684916613,
"loss": 0.8196,
"step": 685
},
{
"epoch": 1.06,
"grad_norm": 1.1816776012865573,
"learning_rate": 0.00016346092244671746,
"loss": 0.8279,
"step": 690
},
{
"epoch": 1.07,
"grad_norm": 1.077779310150086,
"learning_rate": 0.00016276746580515218,
"loss": 0.8997,
"step": 695
},
{
"epoch": 1.07,
"grad_norm": 1.2009550281356633,
"learning_rate": 0.00016206899235035702,
"loss": 0.8938,
"step": 700
},
{
"epoch": 1.08,
"grad_norm": 1.0925262506944138,
"learning_rate": 0.00016136555790919748,
"loss": 0.8856,
"step": 705
},
{
"epoch": 1.09,
"grad_norm": 1.0198591870033458,
"learning_rate": 0.0001606572187050556,
"loss": 0.8978,
"step": 710
},
{
"epoch": 1.1,
"grad_norm": 1.1119975974231693,
"learning_rate": 0.0001599440313533363,
"loss": 0.8876,
"step": 715
},
{
"epoch": 1.11,
"grad_norm": 1.190419344020418,
"learning_rate": 0.00015922605285694215,
"loss": 0.904,
"step": 720
},
{
"epoch": 1.11,
"grad_norm": 2.1639792062638574,
"learning_rate": 0.0001585033406017175,
"loss": 0.9021,
"step": 725
},
{
"epoch": 1.12,
"grad_norm": 1.2040942925245026,
"learning_rate": 0.0001577759523518616,
"loss": 0.9092,
"step": 730
},
{
"epoch": 1.13,
"grad_norm": 1.0704322799863495,
"learning_rate": 0.00015704394624531184,
"loss": 0.8274,
"step": 735
},
{
"epoch": 1.14,
"grad_norm": 1.0627822260269795,
"learning_rate": 0.00015630738078909685,
"loss": 0.8776,
"step": 740
},
{
"epoch": 1.14,
"grad_norm": 0.9905597794182393,
"learning_rate": 0.00015556631485466027,
"loss": 0.9246,
"step": 745
},
{
"epoch": 1.15,
"grad_norm": 1.173456919024251,
"learning_rate": 0.00015482080767315528,
"loss": 0.9656,
"step": 750
},
{
"epoch": 1.16,
"grad_norm": 1.0873012836166025,
"learning_rate": 0.00015407091883071054,
"loss": 0.9464,
"step": 755
},
{
"epoch": 1.17,
"grad_norm": 1.1473043693064626,
"learning_rate": 0.00015331670826366754,
"loss": 0.8496,
"step": 760
},
{
"epoch": 1.17,
"grad_norm": 0.985606297895901,
"learning_rate": 0.00015255823625379017,
"loss": 0.853,
"step": 765
},
{
"epoch": 1.18,
"grad_norm": 1.231267953892478,
"learning_rate": 0.00015179556342344644,
"loss": 0.8652,
"step": 770
},
{
"epoch": 1.19,
"grad_norm": 1.0503811328789359,
"learning_rate": 0.00015102875073076324,
"loss": 0.9447,
"step": 775
},
{
"epoch": 1.2,
"grad_norm": 1.0561678099502803,
"learning_rate": 0.00015025785946475408,
"loss": 0.879,
"step": 780
},
{
"epoch": 1.2,
"grad_norm": 0.9127623837677151,
"learning_rate": 0.00014948295124042057,
"loss": 0.9144,
"step": 785
},
{
"epoch": 1.21,
"grad_norm": 1.147282495230838,
"learning_rate": 0.00014870408799382752,
"loss": 0.9404,
"step": 790
},
{
"epoch": 1.22,
"grad_norm": 1.0608780257484842,
"learning_rate": 0.00014792133197715266,
"loss": 0.9021,
"step": 795
},
{
"epoch": 1.23,
"grad_norm": 1.2163076965365305,
"learning_rate": 0.0001471347457537111,
"loss": 0.915,
"step": 800
},
{
"epoch": 1.24,
"grad_norm": 1.1177869922249701,
"learning_rate": 0.00014634439219295478,
"loss": 0.8648,
"step": 805
},
{
"epoch": 1.24,
"grad_norm": 1.194327396103283,
"learning_rate": 0.0001455503344654474,
"loss": 0.9526,
"step": 810
},
{
"epoch": 1.25,
"grad_norm": 1.0905268007576805,
"learning_rate": 0.00014475263603781554,
"loss": 0.8757,
"step": 815
},
{
"epoch": 1.26,
"grad_norm": 0.9152140993127001,
"learning_rate": 0.0001439513606676759,
"loss": 0.8722,
"step": 820
},
{
"epoch": 1.27,
"grad_norm": 0.9900853991789728,
"learning_rate": 0.00014314657239853927,
"loss": 0.8669,
"step": 825
},
{
"epoch": 1.27,
"grad_norm": 1.0767926512086485,
"learning_rate": 0.000142338335554692,
"loss": 0.8841,
"step": 830
},
{
"epoch": 1.28,
"grad_norm": 0.8724281805811482,
"learning_rate": 0.00014152671473605428,
"loss": 0.826,
"step": 835
},
{
"epoch": 1.29,
"grad_norm": 0.9929237657179188,
"learning_rate": 0.0001407117748130174,
"loss": 0.8765,
"step": 840
},
{
"epoch": 1.3,
"grad_norm": 1.0329119732814056,
"learning_rate": 0.00013989358092125843,
"loss": 0.879,
"step": 845
},
{
"epoch": 1.3,
"grad_norm": 0.8699115976888,
"learning_rate": 0.00013907219845653442,
"loss": 0.8871,
"step": 850
},
{
"epoch": 1.31,
"grad_norm": 2.0311070826655615,
"learning_rate": 0.00013824769306945532,
"loss": 0.9038,
"step": 855
},
{
"epoch": 1.32,
"grad_norm": 1.2265447591069667,
"learning_rate": 0.00013742013066023678,
"loss": 0.8918,
"step": 860
},
{
"epoch": 1.33,
"grad_norm": 1.2487739691504016,
"learning_rate": 0.00013658957737343298,
"loss": 0.8986,
"step": 865
},
{
"epoch": 1.34,
"grad_norm": 0.9999833105305884,
"learning_rate": 0.00013575609959264994,
"loss": 0.9054,
"step": 870
},
{
"epoch": 1.34,
"grad_norm": 1.0265246633147944,
"learning_rate": 0.0001349197639352395,
"loss": 0.8781,
"step": 875
},
{
"epoch": 1.35,
"grad_norm": 1.2752835605974824,
"learning_rate": 0.00013408063724697499,
"loss": 0.9096,
"step": 880
},
{
"epoch": 1.36,
"grad_norm": 1.0841900048671298,
"learning_rate": 0.00013323878659670836,
"loss": 0.8954,
"step": 885
},
{
"epoch": 1.37,
"grad_norm": 0.9822221386317229,
"learning_rate": 0.00013239427927100964,
"loss": 0.9197,
"step": 890
},
{
"epoch": 1.37,
"grad_norm": 1.0664898243823369,
"learning_rate": 0.00013154718276878872,
"loss": 0.8101,
"step": 895
},
{
"epoch": 1.38,
"grad_norm": 0.9728647259678721,
"learning_rate": 0.00013069756479590065,
"loss": 0.8808,
"step": 900
},
{
"epoch": 1.39,
"grad_norm": 1.233908623068181,
"learning_rate": 0.00012984549325973394,
"loss": 0.8942,
"step": 905
},
{
"epoch": 1.4,
"grad_norm": 1.0507024710570716,
"learning_rate": 0.000128991036263783,
"loss": 0.8719,
"step": 910
},
{
"epoch": 1.4,
"grad_norm": 0.9728551705551435,
"learning_rate": 0.0001281342621022048,
"loss": 0.8734,
"step": 915
},
{
"epoch": 1.41,
"grad_norm": 0.9127829023580948,
"learning_rate": 0.00012727523925436026,
"loss": 0.8641,
"step": 920
},
{
"epoch": 1.42,
"grad_norm": 1.114858935845165,
"learning_rate": 0.00012641403637934112,
"loss": 0.8989,
"step": 925
},
{
"epoch": 1.43,
"grad_norm": 1.0021048323600934,
"learning_rate": 0.00012555072231048192,
"loss": 0.8757,
"step": 930
},
{
"epoch": 1.44,
"grad_norm": 0.9850639243326992,
"learning_rate": 0.00012468536604985867,
"loss": 0.8595,
"step": 935
},
{
"epoch": 1.44,
"grad_norm": 0.8816979962904702,
"learning_rate": 0.00012381803676277345,
"loss": 0.8854,
"step": 940
},
{
"epoch": 1.45,
"grad_norm": 1.052381113604825,
"learning_rate": 0.00012294880377222649,
"loss": 0.8966,
"step": 945
},
{
"epoch": 1.46,
"grad_norm": 0.9471916607767124,
"learning_rate": 0.0001220777365533751,
"loss": 0.8977,
"step": 950
},
{
"epoch": 1.47,
"grad_norm": 0.9648558401941475,
"learning_rate": 0.00012120490472798112,
"loss": 0.8837,
"step": 955
},
{
"epoch": 1.47,
"grad_norm": 0.9430401800136761,
"learning_rate": 0.0001203303780588458,
"loss": 0.9009,
"step": 960
},
{
"epoch": 1.48,
"grad_norm": 0.9624197502185033,
"learning_rate": 0.00011945422644423425,
"loss": 0.8645,
"step": 965
},
{
"epoch": 1.49,
"grad_norm": 0.9260810941977449,
"learning_rate": 0.00011857651991228855,
"loss": 0.8243,
"step": 970
},
{
"epoch": 1.5,
"grad_norm": 0.9362374564085868,
"learning_rate": 0.00011769732861543057,
"loss": 0.886,
"step": 975
},
{
"epoch": 1.5,
"grad_norm": 0.9293220698915177,
"learning_rate": 0.00011681672282475495,
"loss": 0.9028,
"step": 980
},
{
"epoch": 1.51,
"grad_norm": 0.9026227086147156,
"learning_rate": 0.00011593477292441251,
"loss": 0.8253,
"step": 985
},
{
"epoch": 1.52,
"grad_norm": 0.9630673700134131,
"learning_rate": 0.00011505154940598468,
"loss": 0.8686,
"step": 990
},
{
"epoch": 1.53,
"grad_norm": 0.8694549872905225,
"learning_rate": 0.00011416712286284943,
"loss": 0.8782,
"step": 995
},
{
"epoch": 1.53,
"grad_norm": 1.007303227993832,
"learning_rate": 0.00011328156398453864,
"loss": 0.8633,
"step": 1000
},
{
"epoch": 1.54,
"grad_norm": 0.9083011875988664,
"learning_rate": 0.00011239494355108848,
"loss": 0.9039,
"step": 1005
},
{
"epoch": 1.55,
"grad_norm": 1.0177463749412752,
"learning_rate": 0.00011150733242738198,
"loss": 0.9029,
"step": 1010
},
{
"epoch": 1.56,
"grad_norm": 0.8631790225704151,
"learning_rate": 0.00011061880155748497,
"loss": 0.8385,
"step": 1015
},
{
"epoch": 1.57,
"grad_norm": 1.5230132007370771,
"learning_rate": 0.00010972942195897582,
"loss": 0.9055,
"step": 1020
},
{
"epoch": 1.57,
"grad_norm": 0.8997463696925398,
"learning_rate": 0.00010883926471726926,
"loss": 0.8656,
"step": 1025
},
{
"epoch": 1.58,
"grad_norm": 0.9377827912335661,
"learning_rate": 0.00010794840097993466,
"loss": 0.9163,
"step": 1030
},
{
"epoch": 1.59,
"grad_norm": 0.9887953595089554,
"learning_rate": 0.00010705690195100939,
"loss": 0.8789,
"step": 1035
},
{
"epoch": 1.6,
"grad_norm": 1.0514798236504352,
"learning_rate": 0.00010616483888530781,
"loss": 0.9027,
"step": 1040
},
{
"epoch": 1.6,
"grad_norm": 0.9297004988976107,
"learning_rate": 0.00010527228308272605,
"loss": 0.9473,
"step": 1045
},
{
"epoch": 1.61,
"grad_norm": 1.079573310483164,
"learning_rate": 0.0001043793058825431,
"loss": 0.8308,
"step": 1050
},
{
"epoch": 1.62,
"grad_norm": 0.8437863342382683,
"learning_rate": 0.00010348597865771909,
"loss": 0.9183,
"step": 1055
},
{
"epoch": 1.63,
"grad_norm": 1.0314459958872142,
"learning_rate": 0.00010259237280919054,
"loss": 0.8965,
"step": 1060
},
{
"epoch": 1.63,
"grad_norm": 1.1300004513750828,
"learning_rate": 0.00010169855976016345,
"loss": 0.9058,
"step": 1065
},
{
"epoch": 1.64,
"grad_norm": 0.915751139105661,
"learning_rate": 0.00010080461095040476,
"loss": 0.8522,
"step": 1070
},
{
"epoch": 1.65,
"grad_norm": 0.9282214710118697,
"learning_rate": 9.991059783053244e-05,
"loss": 0.893,
"step": 1075
},
{
"epoch": 1.66,
"grad_norm": 0.9267500154050946,
"learning_rate": 9.901659185630445e-05,
"loss": 0.9187,
"step": 1080
},
{
"epoch": 1.67,
"grad_norm": 0.9851835947209288,
"learning_rate": 9.812266448290767e-05,
"loss": 0.8489,
"step": 1085
},
{
"epoch": 1.67,
"grad_norm": 0.9072684185548759,
"learning_rate": 9.722888715924664e-05,
"loss": 0.8598,
"step": 1090
},
{
"epoch": 1.68,
"grad_norm": 1.0793886242538702,
"learning_rate": 9.633533132223293e-05,
"loss": 0.9136,
"step": 1095
},
{
"epoch": 1.69,
"grad_norm": 0.9100627615470746,
"learning_rate": 9.54420683910753e-05,
"loss": 0.894,
"step": 1100
},
{
"epoch": 1.7,
"grad_norm": 1.053250149548252,
"learning_rate": 9.454916976157144e-05,
"loss": 0.8604,
"step": 1105
},
{
"epoch": 1.7,
"grad_norm": 1.054945196161409,
"learning_rate": 9.365670680040157e-05,
"loss": 0.8875,
"step": 1110
},
{
"epoch": 1.71,
"grad_norm": 1.0344061549877617,
"learning_rate": 9.276475083942416e-05,
"loss": 0.8612,
"step": 1115
},
{
"epoch": 1.72,
"grad_norm": 1.034253018495178,
"learning_rate": 9.187337316997476e-05,
"loss": 0.8901,
"step": 1120
},
{
"epoch": 1.73,
"grad_norm": 0.9350006681122276,
"learning_rate": 9.09826450371678e-05,
"loss": 0.8913,
"step": 1125
},
{
"epoch": 1.73,
"grad_norm": 1.1854492776108583,
"learning_rate": 9.009263763420228e-05,
"loss": 0.9029,
"step": 1130
},
{
"epoch": 1.74,
"grad_norm": 1.0313744595376304,
"learning_rate": 8.920342209667136e-05,
"loss": 0.8399,
"step": 1135
},
{
"epoch": 1.75,
"grad_norm": 1.109169472722535,
"learning_rate": 8.831506949687685e-05,
"loss": 0.8517,
"step": 1140
},
{
"epoch": 1.76,
"grad_norm": 0.9979711332157661,
"learning_rate": 8.74276508381486e-05,
"loss": 0.8773,
"step": 1145
},
{
"epoch": 1.77,
"grad_norm": 0.9829724669057092,
"learning_rate": 8.654123704916927e-05,
"loss": 0.879,
"step": 1150
},
{
"epoch": 1.77,
"grad_norm": 0.99994654751558,
"learning_rate": 8.565589897830543e-05,
"loss": 0.8523,
"step": 1155
},
{
"epoch": 1.78,
"grad_norm": 0.9478833521675312,
"learning_rate": 8.47717073879447e-05,
"loss": 0.8541,
"step": 1160
},
{
"epoch": 1.79,
"grad_norm": 0.9034620075081865,
"learning_rate": 8.388873294884e-05,
"loss": 0.8918,
"step": 1165
},
{
"epoch": 1.8,
"grad_norm": 0.8723934369993729,
"learning_rate": 8.300704623446111e-05,
"loss": 0.9133,
"step": 1170
},
{
"epoch": 1.8,
"grad_norm": 0.9525657563232377,
"learning_rate": 8.212671771535379e-05,
"loss": 0.8997,
"step": 1175
},
{
"epoch": 1.81,
"grad_norm": 0.9812331418181283,
"learning_rate": 8.124781775350741e-05,
"loss": 0.8877,
"step": 1180
},
{
"epoch": 1.82,
"grad_norm": 1.0880452788188242,
"learning_rate": 8.037041659673105e-05,
"loss": 0.9202,
"step": 1185
},
{
"epoch": 1.83,
"grad_norm": 1.097063406936711,
"learning_rate": 7.949458437303891e-05,
"loss": 0.9068,
"step": 1190
},
{
"epoch": 1.83,
"grad_norm": 0.816681343841031,
"learning_rate": 7.862039108504513e-05,
"loss": 0.8774,
"step": 1195
},
{
"epoch": 1.84,
"grad_norm": 1.0107773305774448,
"learning_rate": 7.774790660436858e-05,
"loss": 0.8973,
"step": 1200
},
{
"epoch": 1.85,
"grad_norm": 1.151927912306129,
"learning_rate": 7.687720066604844e-05,
"loss": 0.8857,
"step": 1205
},
{
"epoch": 1.86,
"grad_norm": 0.7574607659244479,
"learning_rate": 7.600834286297035e-05,
"loss": 0.8681,
"step": 1210
},
{
"epoch": 1.86,
"grad_norm": 0.993971472764409,
"learning_rate": 7.514140264030413e-05,
"loss": 0.9421,
"step": 1215
},
{
"epoch": 1.87,
"grad_norm": 1.1317554618491943,
"learning_rate": 7.427644928995326e-05,
"loss": 0.9151,
"step": 1220
},
{
"epoch": 1.88,
"grad_norm": 0.9946380695782382,
"learning_rate": 7.341355194501638e-05,
"loss": 0.9331,
"step": 1225
},
{
"epoch": 1.89,
"grad_norm": 0.9307204785839222,
"learning_rate": 7.2552779574262e-05,
"loss": 0.9324,
"step": 1230
},
{
"epoch": 1.9,
"grad_norm": 0.8165240509268185,
"learning_rate": 7.16942009766159e-05,
"loss": 0.8972,
"step": 1235
},
{
"epoch": 1.9,
"grad_norm": 1.0160286848183477,
"learning_rate": 7.083788477566206e-05,
"loss": 0.888,
"step": 1240
},
{
"epoch": 1.91,
"grad_norm": 0.9588077943326088,
"learning_rate": 6.998389941415811e-05,
"loss": 0.8776,
"step": 1245
},
{
"epoch": 1.92,
"grad_norm": 1.0027507311971422,
"learning_rate": 6.913231314856467e-05,
"loss": 0.892,
"step": 1250
},
{
"epoch": 1.93,
"grad_norm": 0.8157424560652131,
"learning_rate": 6.828319404358998e-05,
"loss": 0.8611,
"step": 1255
},
{
"epoch": 1.93,
"grad_norm": 0.9346949667365426,
"learning_rate": 6.74366099667495e-05,
"loss": 0.9041,
"step": 1260
},
{
"epoch": 1.94,
"grad_norm": 0.9310369545608014,
"learning_rate": 6.659262858294167e-05,
"loss": 0.8348,
"step": 1265
},
{
"epoch": 1.95,
"grad_norm": 0.916679588029739,
"learning_rate": 6.575131734903952e-05,
"loss": 0.8665,
"step": 1270
},
{
"epoch": 1.96,
"grad_norm": 0.8335818591855719,
"learning_rate": 6.491274350849914e-05,
"loss": 0.8892,
"step": 1275
},
{
"epoch": 1.96,
"grad_norm": 0.9831047716703826,
"learning_rate": 6.407697408598497e-05,
"loss": 0.8944,
"step": 1280
},
{
"epoch": 1.97,
"grad_norm": 0.9268335345851625,
"learning_rate": 6.324407588201292e-05,
"loss": 0.8536,
"step": 1285
},
{
"epoch": 1.98,
"grad_norm": 1.0086597114412514,
"learning_rate": 6.241411546761109e-05,
"loss": 0.8983,
"step": 1290
},
{
"epoch": 1.99,
"grad_norm": 0.9455374773689034,
"learning_rate": 6.158715917899893e-05,
"loss": 0.8638,
"step": 1295
},
{
"epoch": 2.0,
"grad_norm": 0.9581394982958279,
"learning_rate": 6.076327311228522e-05,
"loss": 0.907,
"step": 1300
},
{
"epoch": 2.0,
"eval_loss": 1.1707841157913208,
"eval_runtime": 252.7994,
"eval_samples_per_second": 9.138,
"eval_steps_per_second": 0.574,
"step": 1303
},
{
"epoch": 2.0,
"grad_norm": 1.0117168856081882,
"learning_rate": 5.99425231181853e-05,
"loss": 0.8658,
"step": 1305
},
{
"epoch": 2.01,
"grad_norm": 0.8313758221887009,
"learning_rate": 5.9124974796757614e-05,
"loss": 0.803,
"step": 1310
},
{
"epoch": 2.02,
"grad_norm": 1.1429742233636946,
"learning_rate": 5.831069349216069e-05,
"loss": 0.854,
"step": 1315
},
{
"epoch": 2.03,
"grad_norm": 1.2005681342348582,
"learning_rate": 5.7499744287430366e-05,
"loss": 0.8209,
"step": 1320
},
{
"epoch": 2.03,
"grad_norm": 1.0882362071151326,
"learning_rate": 5.6692191999277614e-05,
"loss": 0.8182,
"step": 1325
},
{
"epoch": 2.04,
"grad_norm": 1.0316332870317524,
"learning_rate": 5.588810117290843e-05,
"loss": 0.888,
"step": 1330
},
{
"epoch": 2.05,
"grad_norm": 0.8415700670907011,
"learning_rate": 5.508753607686452e-05,
"loss": 0.8274,
"step": 1335
},
{
"epoch": 2.06,
"grad_norm": 1.19035695841182,
"learning_rate": 5.429056069788663e-05,
"loss": 0.8587,
"step": 1340
},
{
"epoch": 2.06,
"grad_norm": 0.8282530747648139,
"learning_rate": 5.3497238735800456e-05,
"loss": 0.8582,
"step": 1345
},
{
"epoch": 2.07,
"grad_norm": 0.8236642419616373,
"learning_rate": 5.2707633598425023e-05,
"loss": 0.8242,
"step": 1350
},
{
"epoch": 2.08,
"grad_norm": 0.9499844942270589,
"learning_rate": 5.192180839650482e-05,
"loss": 0.8419,
"step": 1355
},
{
"epoch": 2.09,
"grad_norm": 0.9665172203162608,
"learning_rate": 5.1139825938665706e-05,
"loss": 0.8168,
"step": 1360
},
{
"epoch": 2.1,
"grad_norm": 0.9570953639965174,
"learning_rate": 5.036174872639443e-05,
"loss": 0.7975,
"step": 1365
},
{
"epoch": 2.1,
"grad_norm": 1.1769715973472896,
"learning_rate": 4.95876389490435e-05,
"loss": 0.8803,
"step": 1370
},
{
"epoch": 2.11,
"grad_norm": 1.0737767072312128,
"learning_rate": 4.8817558478860316e-05,
"loss": 0.8392,
"step": 1375
},
{
"epoch": 2.12,
"grad_norm": 1.0356976294210105,
"learning_rate": 4.805156886604192e-05,
"loss": 0.8427,
"step": 1380
},
{
"epoch": 2.13,
"grad_norm": 0.9731310972525462,
"learning_rate": 4.728973133381557e-05,
"loss": 0.8422,
"step": 1385
},
{
"epoch": 2.13,
"grad_norm": 1.0915183222326856,
"learning_rate": 4.6532106773545356e-05,
"loss": 0.8002,
"step": 1390
},
{
"epoch": 2.14,
"grad_norm": 1.0419392375432739,
"learning_rate": 4.5778755739865234e-05,
"loss": 0.8035,
"step": 1395
},
{
"epoch": 2.15,
"grad_norm": 1.034845804276428,
"learning_rate": 4.5029738445839143e-05,
"loss": 0.8633,
"step": 1400
},
{
"epoch": 2.16,
"grad_norm": 0.9808221912256501,
"learning_rate": 4.4285114758148385e-05,
"loss": 0.8238,
"step": 1405
},
{
"epoch": 2.16,
"grad_norm": 0.9518518180660357,
"learning_rate": 4.3544944192306536e-05,
"loss": 0.821,
"step": 1410
},
{
"epoch": 2.17,
"grad_norm": 1.0002459852548866,
"learning_rate": 4.2809285907902804e-05,
"loss": 0.8153,
"step": 1415
},
{
"epoch": 2.18,
"grad_norm": 0.9090185258143434,
"learning_rate": 4.207819870387331e-05,
"loss": 0.8724,
"step": 1420
},
{
"epoch": 2.19,
"grad_norm": 0.8794828030312869,
"learning_rate": 4.135174101380154e-05,
"loss": 0.7995,
"step": 1425
},
{
"epoch": 2.19,
"grad_norm": 0.8426944935252626,
"learning_rate": 4.0629970901248125e-05,
"loss": 0.878,
"step": 1430
},
{
"epoch": 2.2,
"grad_norm": 1.0954146157200644,
"learning_rate": 3.991294605510969e-05,
"loss": 0.8605,
"step": 1435
},
{
"epoch": 2.21,
"grad_norm": 1.0887405499363092,
"learning_rate": 3.920072378500814e-05,
"loss": 0.853,
"step": 1440
},
{
"epoch": 2.22,
"grad_norm": 1.0547395334098018,
"learning_rate": 3.849336101671015e-05,
"loss": 0.8921,
"step": 1445
},
{
"epoch": 2.23,
"grad_norm": 1.069888150135721,
"learning_rate": 3.779091428757692e-05,
"loss": 0.8161,
"step": 1450
},
{
"epoch": 2.23,
"grad_norm": 0.9353341753814841,
"learning_rate": 3.709343974204577e-05,
"loss": 0.8179,
"step": 1455
},
{
"epoch": 2.24,
"grad_norm": 1.0108851490383193,
"learning_rate": 3.640099312714235e-05,
"loss": 0.8385,
"step": 1460
},
{
"epoch": 2.25,
"grad_norm": 1.064940293841458,
"learning_rate": 3.5713629788025036e-05,
"loss": 0.8135,
"step": 1465
},
{
"epoch": 2.26,
"grad_norm": 1.1487280639099855,
"learning_rate": 3.503140466356151e-05,
"loss": 0.8021,
"step": 1470
},
{
"epoch": 2.26,
"grad_norm": 0.9701769325400604,
"learning_rate": 3.435437228193741e-05,
"loss": 0.8399,
"step": 1475
},
{
"epoch": 2.27,
"grad_norm": 0.9942305006661968,
"learning_rate": 3.3682586756298185e-05,
"loss": 0.8427,
"step": 1480
},
{
"epoch": 2.28,
"grad_norm": 1.0099321403910597,
"learning_rate": 3.3016101780424146e-05,
"loss": 0.8693,
"step": 1485
},
{
"epoch": 2.29,
"grad_norm": 1.169305983316981,
"learning_rate": 3.235497062443852e-05,
"loss": 0.8545,
"step": 1490
},
{
"epoch": 2.29,
"grad_norm": 0.9289433553993577,
"learning_rate": 3.169924613055003e-05,
"loss": 0.7912,
"step": 1495
},
{
"epoch": 2.3,
"grad_norm": 1.1184033184773026,
"learning_rate": 3.10489807088294e-05,
"loss": 0.8488,
"step": 1500
},
{
"epoch": 2.31,
"grad_norm": 0.8651320149450171,
"learning_rate": 3.0404226333020114e-05,
"loss": 0.8063,
"step": 1505
},
{
"epoch": 2.32,
"grad_norm": 1.015380797479418,
"learning_rate": 2.976503453638452e-05,
"loss": 0.8281,
"step": 1510
},
{
"epoch": 2.33,
"grad_norm": 0.9105715510091448,
"learning_rate": 2.9131456407584912e-05,
"loss": 0.8396,
"step": 1515
},
{
"epoch": 2.33,
"grad_norm": 0.8953944449056994,
"learning_rate": 2.8503542586600095e-05,
"loss": 0.8059,
"step": 1520
},
{
"epoch": 2.34,
"grad_norm": 0.9445583773531011,
"learning_rate": 2.7881343260677938e-05,
"loss": 0.7555,
"step": 1525
},
{
"epoch": 2.35,
"grad_norm": 0.9760863299166253,
"learning_rate": 2.7264908160324044e-05,
"loss": 0.872,
"step": 1530
},
{
"epoch": 2.36,
"grad_norm": 0.9217062841634305,
"learning_rate": 2.66542865553269e-05,
"loss": 0.8772,
"step": 1535
},
{
"epoch": 2.36,
"grad_norm": 0.9862000270745448,
"learning_rate": 2.6049527250820048e-05,
"loss": 0.8042,
"step": 1540
},
{
"epoch": 2.37,
"grad_norm": 0.9372789886762245,
"learning_rate": 2.5450678583381037e-05,
"loss": 0.8373,
"step": 1545
},
{
"epoch": 2.38,
"grad_norm": 0.9849583322670422,
"learning_rate": 2.4857788417168082e-05,
"loss": 0.8449,
"step": 1550
},
{
"epoch": 2.39,
"grad_norm": 0.9531252163149414,
"learning_rate": 2.4270904140094597e-05,
"loss": 0.8204,
"step": 1555
},
{
"epoch": 2.39,
"grad_norm": 1.1160019622624047,
"learning_rate": 2.3690072660041373e-05,
"loss": 0.857,
"step": 1560
},
{
"epoch": 2.4,
"grad_norm": 1.0574865907973916,
"learning_rate": 2.3115340401107487e-05,
"loss": 0.8154,
"step": 1565
},
{
"epoch": 2.41,
"grad_norm": 1.0014445052570997,
"learning_rate": 2.254675329989988e-05,
"loss": 0.8526,
"step": 1570
},
{
"epoch": 2.42,
"grad_norm": 1.1337662282318417,
"learning_rate": 2.1984356801861506e-05,
"loss": 0.7529,
"step": 1575
},
{
"epoch": 2.43,
"grad_norm": 1.0240125708817747,
"learning_rate": 2.1428195857639256e-05,
"loss": 0.8252,
"step": 1580
},
{
"epoch": 2.43,
"grad_norm": 0.8777879796341344,
"learning_rate": 2.0878314919491183e-05,
"loss": 0.8349,
"step": 1585
},
{
"epoch": 2.44,
"grad_norm": 0.9749914994991105,
"learning_rate": 2.0334757937733374e-05,
"loss": 0.8147,
"step": 1590
},
{
"epoch": 2.45,
"grad_norm": 1.0047182891795483,
"learning_rate": 1.9797568357227293e-05,
"loss": 0.8225,
"step": 1595
},
{
"epoch": 2.46,
"grad_norm": 1.102505598943093,
"learning_rate": 1.92667891139074e-05,
"loss": 0.8356,
"step": 1600
},
{
"epoch": 2.46,
"grad_norm": 0.9052644924404957,
"learning_rate": 1.8742462631349246e-05,
"loss": 0.8509,
"step": 1605
},
{
"epoch": 2.47,
"grad_norm": 0.9927305294402076,
"learning_rate": 1.822463081737883e-05,
"loss": 0.8182,
"step": 1610
},
{
"epoch": 2.48,
"grad_norm": 1.0563552318088962,
"learning_rate": 1.7713335060722946e-05,
"loss": 0.7578,
"step": 1615
},
{
"epoch": 2.49,
"grad_norm": 1.0170016112678426,
"learning_rate": 1.720861622770116e-05,
"loss": 0.7562,
"step": 1620
},
{
"epoch": 2.49,
"grad_norm": 0.9404587306406071,
"learning_rate": 1.671051465895953e-05,
"loss": 0.786,
"step": 1625
},
{
"epoch": 2.5,
"grad_norm": 1.0171241452919628,
"learning_rate": 1.6219070166246154e-05,
"loss": 0.8616,
"step": 1630
},
{
"epoch": 2.51,
"grad_norm": 1.0322115593813674,
"learning_rate": 1.5734322029229253e-05,
"loss": 0.8592,
"step": 1635
},
{
"epoch": 2.52,
"grad_norm": 1.0192391783063095,
"learning_rate": 1.5256308992357716e-05,
"loss": 0.8372,
"step": 1640
},
{
"epoch": 2.52,
"grad_norm": 1.17212669377524,
"learning_rate": 1.4785069261764184e-05,
"loss": 0.7713,
"step": 1645
},
{
"epoch": 2.53,
"grad_norm": 1.0439043355220827,
"learning_rate": 1.4320640502211536e-05,
"loss": 0.8379,
"step": 1650
},
{
"epoch": 2.54,
"grad_norm": 0.9140306646774222,
"learning_rate": 1.386305983408236e-05,
"loss": 0.8041,
"step": 1655
},
{
"epoch": 2.55,
"grad_norm": 1.061796317725875,
"learning_rate": 1.3412363830412078e-05,
"loss": 0.8318,
"step": 1660
},
{
"epoch": 2.56,
"grad_norm": 1.0834691203563807,
"learning_rate": 1.2968588513965706e-05,
"loss": 0.851,
"step": 1665
},
{
"epoch": 2.56,
"grad_norm": 0.9887481512630522,
"learning_rate": 1.2531769354358825e-05,
"loss": 0.8728,
"step": 1670
},
{
"epoch": 2.57,
"grad_norm": 1.058272033476656,
"learning_rate": 1.2101941265222373e-05,
"loss": 0.8016,
"step": 1675
},
{
"epoch": 2.58,
"grad_norm": 1.072187651154863,
"learning_rate": 1.1679138601412255e-05,
"loss": 0.8505,
"step": 1680
},
{
"epoch": 2.59,
"grad_norm": 1.2017961960493708,
"learning_rate": 1.126339515626349e-05,
"loss": 0.8601,
"step": 1685
},
{
"epoch": 2.59,
"grad_norm": 0.9334846571871207,
"learning_rate": 1.0854744158889085e-05,
"loss": 0.8178,
"step": 1690
},
{
"epoch": 2.6,
"grad_norm": 0.9385801565830063,
"learning_rate": 1.0453218271524224e-05,
"loss": 0.8155,
"step": 1695
},
{
"epoch": 2.61,
"grad_norm": 1.0026372920490085,
"learning_rate": 1.0058849586915653e-05,
"loss": 0.8463,
"step": 1700
},
{
"epoch": 2.62,
"grad_norm": 1.0094111622202961,
"learning_rate": 9.671669625756574e-06,
"loss": 0.8291,
"step": 1705
},
{
"epoch": 2.62,
"grad_norm": 1.0498204906448678,
"learning_rate": 9.291709334167397e-06,
"loss": 0.8694,
"step": 1710
},
{
"epoch": 2.63,
"grad_norm": 0.9108848375007711,
"learning_rate": 8.918999081222156e-06,
"loss": 0.8154,
"step": 1715
},
{
"epoch": 2.64,
"grad_norm": 1.1045628537182532,
"learning_rate": 8.553568656521293e-06,
"loss": 0.8542,
"step": 1720
},
{
"epoch": 2.65,
"grad_norm": 0.9187751707089163,
"learning_rate": 8.195447267810686e-06,
"loss": 0.8847,
"step": 1725
},
{
"epoch": 2.66,
"grad_norm": 1.0976708579458017,
"learning_rate": 7.844663538647101e-06,
"loss": 0.8168,
"step": 1730
},
{
"epoch": 2.66,
"grad_norm": 1.0277757622039416,
"learning_rate": 7.501245506110433e-06,
"loss": 0.824,
"step": 1735
},
{
"epoch": 2.67,
"grad_norm": 0.9800736010664038,
"learning_rate": 7.165220618562751e-06,
"loss": 0.8499,
"step": 1740
},
{
"epoch": 2.68,
"grad_norm": 1.1066605466255004,
"learning_rate": 6.83661573345451e-06,
"loss": 0.869,
"step": 1745
},
{
"epoch": 2.69,
"grad_norm": 1.0217949991902664,
"learning_rate": 6.515457115177803e-06,
"loss": 0.8308,
"step": 1750
},
{
"epoch": 2.69,
"grad_norm": 1.0285372086807927,
"learning_rate": 6.20177043296728e-06,
"loss": 0.8513,
"step": 1755
},
{
"epoch": 2.7,
"grad_norm": 0.9228807374530837,
"learning_rate": 5.895580758848318e-06,
"loss": 0.8359,
"step": 1760
},
{
"epoch": 2.71,
"grad_norm": 1.042562235261924,
"learning_rate": 5.596912565633184e-06,
"loss": 0.8144,
"step": 1765
},
{
"epoch": 2.72,
"grad_norm": 1.098352136508281,
"learning_rate": 5.305789724965038e-06,
"loss": 0.8253,
"step": 1770
},
{
"epoch": 2.72,
"grad_norm": 0.9667182261715308,
"learning_rate": 5.022235505409823e-06,
"loss": 0.8672,
"step": 1775
},
{
"epoch": 2.73,
"grad_norm": 1.019562250529469,
"learning_rate": 4.746272570596555e-06,
"loss": 0.8214,
"step": 1780
},
{
"epoch": 2.74,
"grad_norm": 1.2161564555157138,
"learning_rate": 4.477922977405913e-06,
"loss": 0.8113,
"step": 1785
},
{
"epoch": 2.75,
"grad_norm": 1.0015900069881538,
"learning_rate": 4.217208174207199e-06,
"loss": 0.8484,
"step": 1790
},
{
"epoch": 2.76,
"grad_norm": 0.9631542498199687,
"learning_rate": 3.964148999144202e-06,
"loss": 0.853,
"step": 1795
},
{
"epoch": 2.76,
"grad_norm": 0.9926634902794105,
"learning_rate": 3.71876567846946e-06,
"loss": 0.8488,
"step": 1800
},
{
"epoch": 2.77,
"grad_norm": 0.8438960246283821,
"learning_rate": 3.481077824927792e-06,
"loss": 0.829,
"step": 1805
},
{
"epoch": 2.78,
"grad_norm": 1.0027991142986536,
"learning_rate": 3.251104436188679e-06,
"loss": 0.8416,
"step": 1810
},
{
"epoch": 2.79,
"grad_norm": 1.0903635867609474,
"learning_rate": 3.0288638933277934e-06,
"loss": 0.8065,
"step": 1815
},
{
"epoch": 2.79,
"grad_norm": 1.0073217221247301,
"learning_rate": 2.8143739593578856e-06,
"loss": 0.7876,
"step": 1820
},
{
"epoch": 2.8,
"grad_norm": 0.8171096137144128,
"learning_rate": 2.607651777809039e-06,
"loss": 0.7574,
"step": 1825
},
{
"epoch": 2.81,
"grad_norm": 0.9374489147765585,
"learning_rate": 2.4087138713584367e-06,
"loss": 0.8652,
"step": 1830
},
{
"epoch": 2.82,
"grad_norm": 1.054601585067846,
"learning_rate": 2.2175761405097584e-06,
"loss": 0.8471,
"step": 1835
},
{
"epoch": 2.82,
"grad_norm": 1.064632352850785,
"learning_rate": 2.0342538623222997e-06,
"loss": 0.8189,
"step": 1840
},
{
"epoch": 2.83,
"grad_norm": 0.8546672110619797,
"learning_rate": 1.8587616891899363e-06,
"loss": 0.7818,
"step": 1845
},
{
"epoch": 2.84,
"grad_norm": 0.9645255792031215,
"learning_rate": 1.6911136476699508e-06,
"loss": 0.7776,
"step": 1850
},
{
"epoch": 2.85,
"grad_norm": 1.068610540574793,
"learning_rate": 1.5313231373619952e-06,
"loss": 0.7857,
"step": 1855
},
{
"epoch": 2.85,
"grad_norm": 1.100889943747955,
"learning_rate": 1.3794029298370814e-06,
"loss": 0.8141,
"step": 1860
},
{
"epoch": 2.86,
"grad_norm": 0.9783722379990912,
"learning_rate": 1.2353651676167643e-06,
"loss": 0.8358,
"step": 1865
},
{
"epoch": 2.87,
"grad_norm": 0.8803200725105156,
"learning_rate": 1.0992213632026517e-06,
"loss": 0.8678,
"step": 1870
},
{
"epoch": 2.88,
"grad_norm": 1.1249544133659088,
"learning_rate": 9.709823981562282e-07,
"loss": 0.8078,
"step": 1875
},
{
"epoch": 2.89,
"grad_norm": 1.056395258554381,
"learning_rate": 8.506585222291752e-07,
"loss": 0.7749,
"step": 1880
},
{
"epoch": 2.89,
"grad_norm": 0.9253933119169012,
"learning_rate": 7.382593525440573e-07,
"loss": 0.8008,
"step": 1885
},
{
"epoch": 2.9,
"grad_norm": 0.9212088582083587,
"learning_rate": 6.337938728257054e-07,
"loss": 0.8333,
"step": 1890
},
{
"epoch": 2.91,
"grad_norm": 1.100818174265163,
"learning_rate": 5.372704326831901e-07,
"loss": 0.8342,
"step": 1895
},
{
"epoch": 2.92,
"grad_norm": 0.8164157006379008,
"learning_rate": 4.486967469424008e-07,
"loss": 0.8583,
"step": 1900
},
{
"epoch": 2.92,
"grad_norm": 1.0674250153511942,
"learning_rate": 3.6807989502949394e-07,
"loss": 0.8213,
"step": 1905
},
{
"epoch": 2.93,
"grad_norm": 0.9095567427952401,
"learning_rate": 2.954263204050123e-07,
"loss": 0.8252,
"step": 1910
},
{
"epoch": 2.94,
"grad_norm": 1.0738638532567681,
"learning_rate": 2.3074183004887505e-07,
"loss": 0.8448,
"step": 1915
},
{
"epoch": 2.95,
"grad_norm": 1.1019761677075681,
"learning_rate": 1.7403159399629332e-07,
"loss": 0.8088,
"step": 1920
},
{
"epoch": 2.95,
"grad_norm": 0.9934974979812884,
"learning_rate": 1.2530014492446728e-07,
"loss": 0.8029,
"step": 1925
},
{
"epoch": 2.96,
"grad_norm": 1.0433381527945313,
"learning_rate": 8.455137779038724e-08,
"loss": 0.8209,
"step": 1930
},
{
"epoch": 2.97,
"grad_norm": 1.032700859857025,
"learning_rate": 5.1788549519438124e-08,
"loss": 0.7365,
"step": 1935
},
{
"epoch": 2.98,
"grad_norm": 0.9408894102132038,
"learning_rate": 2.7014278745163268e-08,
"loss": 0.8451,
"step": 1940
},
{
"epoch": 2.99,
"grad_norm": 1.0657773929136594,
"learning_rate": 1.0230545599909658e-08,
"loss": 0.8246,
"step": 1945
},
{
"epoch": 2.99,
"grad_norm": 1.0107360164317165,
"learning_rate": 1.438691556565619e-09,
"loss": 0.8209,
"step": 1950
},
{
"epoch": 3.0,
"eval_loss": 1.1822351217269897,
"eval_runtime": 252.7889,
"eval_samples_per_second": 9.138,
"eval_steps_per_second": 0.574,
"step": 1953
},
{
"epoch": 3.0,
"step": 1953,
"total_flos": 4109935435055104.0,
"train_loss": 1.1336081770219621,
"train_runtime": 21652.8936,
"train_samples_per_second": 2.888,
"train_steps_per_second": 0.09
}
],
"logging_steps": 5,
"max_steps": 1953,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 4109935435055104.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}