gemma2b-summarize-gpt4o-64k / trainer_state.json
chansung's picture
Model save
2d3c737 verified
raw
history blame
76.4 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 15.0,
"eval_steps": 500,
"global_step": 2190,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.00684931506849315,
"grad_norm": 2.453125,
"learning_rate": 9.132420091324201e-07,
"loss": 3.0017,
"step": 1
},
{
"epoch": 0.03424657534246575,
"grad_norm": 3.34375,
"learning_rate": 4.566210045662101e-06,
"loss": 3.0722,
"step": 5
},
{
"epoch": 0.0684931506849315,
"grad_norm": 2.359375,
"learning_rate": 9.132420091324201e-06,
"loss": 3.0404,
"step": 10
},
{
"epoch": 0.10273972602739725,
"grad_norm": 1.7890625,
"learning_rate": 1.3698630136986302e-05,
"loss": 3.0338,
"step": 15
},
{
"epoch": 0.136986301369863,
"grad_norm": 2.421875,
"learning_rate": 1.8264840182648402e-05,
"loss": 3.023,
"step": 20
},
{
"epoch": 0.17123287671232876,
"grad_norm": 14.125,
"learning_rate": 2.2831050228310503e-05,
"loss": 2.9441,
"step": 25
},
{
"epoch": 0.2054794520547945,
"grad_norm": 2.609375,
"learning_rate": 2.7397260273972603e-05,
"loss": 2.8289,
"step": 30
},
{
"epoch": 0.23972602739726026,
"grad_norm": 1.3203125,
"learning_rate": 3.1963470319634704e-05,
"loss": 2.6567,
"step": 35
},
{
"epoch": 0.273972602739726,
"grad_norm": 1.5703125,
"learning_rate": 3.6529680365296805e-05,
"loss": 2.5383,
"step": 40
},
{
"epoch": 0.3082191780821918,
"grad_norm": 1.8671875,
"learning_rate": 4.1095890410958905e-05,
"loss": 2.4484,
"step": 45
},
{
"epoch": 0.3424657534246575,
"grad_norm": 1.140625,
"learning_rate": 4.5662100456621006e-05,
"loss": 2.3211,
"step": 50
},
{
"epoch": 0.3767123287671233,
"grad_norm": 1.6640625,
"learning_rate": 5.0228310502283106e-05,
"loss": 2.1832,
"step": 55
},
{
"epoch": 0.410958904109589,
"grad_norm": 82.0,
"learning_rate": 5.479452054794521e-05,
"loss": 2.0896,
"step": 60
},
{
"epoch": 0.4452054794520548,
"grad_norm": 1.7890625,
"learning_rate": 5.936073059360731e-05,
"loss": 1.9847,
"step": 65
},
{
"epoch": 0.4794520547945205,
"grad_norm": 0.61328125,
"learning_rate": 6.392694063926941e-05,
"loss": 1.8803,
"step": 70
},
{
"epoch": 0.5136986301369864,
"grad_norm": 0.7890625,
"learning_rate": 6.84931506849315e-05,
"loss": 1.7981,
"step": 75
},
{
"epoch": 0.547945205479452,
"grad_norm": 0.474609375,
"learning_rate": 7.305936073059361e-05,
"loss": 1.7173,
"step": 80
},
{
"epoch": 0.5821917808219178,
"grad_norm": 0.4609375,
"learning_rate": 7.76255707762557e-05,
"loss": 1.6268,
"step": 85
},
{
"epoch": 0.6164383561643836,
"grad_norm": 0.455078125,
"learning_rate": 8.219178082191781e-05,
"loss": 1.5857,
"step": 90
},
{
"epoch": 0.6506849315068494,
"grad_norm": 0.298828125,
"learning_rate": 8.67579908675799e-05,
"loss": 1.5367,
"step": 95
},
{
"epoch": 0.684931506849315,
"grad_norm": 0.37109375,
"learning_rate": 9.132420091324201e-05,
"loss": 1.4903,
"step": 100
},
{
"epoch": 0.7191780821917808,
"grad_norm": 0.31640625,
"learning_rate": 9.58904109589041e-05,
"loss": 1.4631,
"step": 105
},
{
"epoch": 0.7534246575342466,
"grad_norm": 0.30859375,
"learning_rate": 0.00010045662100456621,
"loss": 1.423,
"step": 110
},
{
"epoch": 0.7876712328767124,
"grad_norm": 0.2490234375,
"learning_rate": 0.00010502283105022832,
"loss": 1.3912,
"step": 115
},
{
"epoch": 0.821917808219178,
"grad_norm": 0.466796875,
"learning_rate": 0.00010958904109589041,
"loss": 1.3723,
"step": 120
},
{
"epoch": 0.8561643835616438,
"grad_norm": 0.2890625,
"learning_rate": 0.00011415525114155252,
"loss": 1.3465,
"step": 125
},
{
"epoch": 0.8904109589041096,
"grad_norm": 0.5625,
"learning_rate": 0.00011872146118721462,
"loss": 1.3393,
"step": 130
},
{
"epoch": 0.9246575342465754,
"grad_norm": 0.271484375,
"learning_rate": 0.0001232876712328767,
"loss": 1.3149,
"step": 135
},
{
"epoch": 0.958904109589041,
"grad_norm": 0.314453125,
"learning_rate": 0.00012785388127853882,
"loss": 1.3162,
"step": 140
},
{
"epoch": 0.9931506849315068,
"grad_norm": 0.412109375,
"learning_rate": 0.00013242009132420092,
"loss": 1.2959,
"step": 145
},
{
"epoch": 1.0,
"eval_loss": 2.529484510421753,
"eval_runtime": 0.5429,
"eval_samples_per_second": 18.42,
"eval_steps_per_second": 1.842,
"step": 146
},
{
"epoch": 1.0273972602739727,
"grad_norm": 0.57421875,
"learning_rate": 0.000136986301369863,
"loss": 1.2853,
"step": 150
},
{
"epoch": 1.0616438356164384,
"grad_norm": 0.50390625,
"learning_rate": 0.0001415525114155251,
"loss": 1.2733,
"step": 155
},
{
"epoch": 1.095890410958904,
"grad_norm": 0.41015625,
"learning_rate": 0.00014611872146118722,
"loss": 1.2646,
"step": 160
},
{
"epoch": 1.13013698630137,
"grad_norm": 0.330078125,
"learning_rate": 0.00015068493150684933,
"loss": 1.2417,
"step": 165
},
{
"epoch": 1.1643835616438356,
"grad_norm": 0.37890625,
"learning_rate": 0.0001552511415525114,
"loss": 1.2469,
"step": 170
},
{
"epoch": 1.1986301369863013,
"grad_norm": 0.32421875,
"learning_rate": 0.00015981735159817351,
"loss": 1.2308,
"step": 175
},
{
"epoch": 1.2328767123287672,
"grad_norm": 0.4375,
"learning_rate": 0.00016438356164383562,
"loss": 1.2253,
"step": 180
},
{
"epoch": 1.2671232876712328,
"grad_norm": 0.60546875,
"learning_rate": 0.00016894977168949773,
"loss": 1.2334,
"step": 185
},
{
"epoch": 1.3013698630136985,
"grad_norm": 0.734375,
"learning_rate": 0.0001735159817351598,
"loss": 1.2207,
"step": 190
},
{
"epoch": 1.3356164383561644,
"grad_norm": 0.3671875,
"learning_rate": 0.00017808219178082192,
"loss": 1.212,
"step": 195
},
{
"epoch": 1.36986301369863,
"grad_norm": 0.37890625,
"learning_rate": 0.00018264840182648402,
"loss": 1.2152,
"step": 200
},
{
"epoch": 1.404109589041096,
"grad_norm": 0.3359375,
"learning_rate": 0.00018721461187214613,
"loss": 1.2071,
"step": 205
},
{
"epoch": 1.4383561643835616,
"grad_norm": 0.361328125,
"learning_rate": 0.0001917808219178082,
"loss": 1.2011,
"step": 210
},
{
"epoch": 1.4726027397260273,
"grad_norm": 0.345703125,
"learning_rate": 0.00019634703196347032,
"loss": 1.2017,
"step": 215
},
{
"epoch": 1.5068493150684932,
"grad_norm": 0.6484375,
"learning_rate": 0.00019999987297289245,
"loss": 1.1886,
"step": 220
},
{
"epoch": 1.541095890410959,
"grad_norm": 0.482421875,
"learning_rate": 0.00019999542705801296,
"loss": 1.1932,
"step": 225
},
{
"epoch": 1.5753424657534247,
"grad_norm": 0.7578125,
"learning_rate": 0.00019998463011046926,
"loss": 1.1888,
"step": 230
},
{
"epoch": 1.6095890410958904,
"grad_norm": 1.5234375,
"learning_rate": 0.00019996748281601038,
"loss": 1.1868,
"step": 235
},
{
"epoch": 1.643835616438356,
"grad_norm": 1.1875,
"learning_rate": 0.00019994398626371643,
"loss": 1.1775,
"step": 240
},
{
"epoch": 1.678082191780822,
"grad_norm": 0.796875,
"learning_rate": 0.0001999141419459293,
"loss": 1.192,
"step": 245
},
{
"epoch": 1.7123287671232876,
"grad_norm": 0.9140625,
"learning_rate": 0.00019987795175815807,
"loss": 1.1756,
"step": 250
},
{
"epoch": 1.7465753424657535,
"grad_norm": 0.337890625,
"learning_rate": 0.0001998354179989585,
"loss": 1.1747,
"step": 255
},
{
"epoch": 1.7808219178082192,
"grad_norm": 0.546875,
"learning_rate": 0.0001997865433697871,
"loss": 1.1785,
"step": 260
},
{
"epoch": 1.8150684931506849,
"grad_norm": 0.5625,
"learning_rate": 0.00019973133097482947,
"loss": 1.158,
"step": 265
},
{
"epoch": 1.8493150684931505,
"grad_norm": 0.439453125,
"learning_rate": 0.00019966978432080316,
"loss": 1.1678,
"step": 270
},
{
"epoch": 1.8835616438356164,
"grad_norm": 0.609375,
"learning_rate": 0.00019960190731673505,
"loss": 1.1646,
"step": 275
},
{
"epoch": 1.9178082191780823,
"grad_norm": 0.3984375,
"learning_rate": 0.00019952770427371304,
"loss": 1.1514,
"step": 280
},
{
"epoch": 1.952054794520548,
"grad_norm": 0.33984375,
"learning_rate": 0.00019944717990461207,
"loss": 1.1478,
"step": 285
},
{
"epoch": 1.9863013698630136,
"grad_norm": 0.5703125,
"learning_rate": 0.00019936033932379504,
"loss": 1.1524,
"step": 290
},
{
"epoch": 2.0,
"eval_loss": 2.491286277770996,
"eval_runtime": 0.5472,
"eval_samples_per_second": 18.275,
"eval_steps_per_second": 1.827,
"step": 292
},
{
"epoch": 2.0205479452054793,
"grad_norm": 0.419921875,
"learning_rate": 0.00019926718804678785,
"loss": 1.1515,
"step": 295
},
{
"epoch": 2.0547945205479454,
"grad_norm": 0.384765625,
"learning_rate": 0.000199167731989929,
"loss": 1.1337,
"step": 300
},
{
"epoch": 2.089041095890411,
"grad_norm": 0.34765625,
"learning_rate": 0.00019906197746999408,
"loss": 1.1328,
"step": 305
},
{
"epoch": 2.1232876712328768,
"grad_norm": 0.36328125,
"learning_rate": 0.00019894993120379435,
"loss": 1.1231,
"step": 310
},
{
"epoch": 2.1575342465753424,
"grad_norm": 0.427734375,
"learning_rate": 0.00019883160030775016,
"loss": 1.1349,
"step": 315
},
{
"epoch": 2.191780821917808,
"grad_norm": 0.35546875,
"learning_rate": 0.00019870699229743911,
"loss": 1.1291,
"step": 320
},
{
"epoch": 2.2260273972602738,
"grad_norm": 0.76171875,
"learning_rate": 0.0001985761150871185,
"loss": 1.1263,
"step": 325
},
{
"epoch": 2.26027397260274,
"grad_norm": 0.71875,
"learning_rate": 0.00019843897698922284,
"loss": 1.1248,
"step": 330
},
{
"epoch": 2.2945205479452055,
"grad_norm": 0.8046875,
"learning_rate": 0.00019829558671383585,
"loss": 1.1254,
"step": 335
},
{
"epoch": 2.328767123287671,
"grad_norm": 0.439453125,
"learning_rate": 0.00019814595336813725,
"loss": 1.118,
"step": 340
},
{
"epoch": 2.363013698630137,
"grad_norm": 0.53125,
"learning_rate": 0.0001979900864558242,
"loss": 1.1152,
"step": 345
},
{
"epoch": 2.3972602739726026,
"grad_norm": 0.376953125,
"learning_rate": 0.00019782799587650805,
"loss": 1.133,
"step": 350
},
{
"epoch": 2.4315068493150687,
"grad_norm": 0.66796875,
"learning_rate": 0.00019765969192508508,
"loss": 1.1186,
"step": 355
},
{
"epoch": 2.4657534246575343,
"grad_norm": 0.671875,
"learning_rate": 0.00019748518529108316,
"loss": 1.1231,
"step": 360
},
{
"epoch": 2.5,
"grad_norm": 0.42578125,
"learning_rate": 0.00019730448705798239,
"loss": 1.1229,
"step": 365
},
{
"epoch": 2.5342465753424657,
"grad_norm": 0.361328125,
"learning_rate": 0.00019711760870251143,
"loss": 1.1169,
"step": 370
},
{
"epoch": 2.5684931506849313,
"grad_norm": 0.41796875,
"learning_rate": 0.00019692456209391846,
"loss": 1.1096,
"step": 375
},
{
"epoch": 2.602739726027397,
"grad_norm": 0.451171875,
"learning_rate": 0.0001967253594932173,
"loss": 1.1121,
"step": 380
},
{
"epoch": 2.636986301369863,
"grad_norm": 0.390625,
"learning_rate": 0.00019652001355240878,
"loss": 1.1203,
"step": 385
},
{
"epoch": 2.671232876712329,
"grad_norm": 0.37109375,
"learning_rate": 0.00019630853731367713,
"loss": 1.1165,
"step": 390
},
{
"epoch": 2.7054794520547945,
"grad_norm": 0.4140625,
"learning_rate": 0.0001960909442085615,
"loss": 1.1158,
"step": 395
},
{
"epoch": 2.73972602739726,
"grad_norm": 0.400390625,
"learning_rate": 0.00019586724805710306,
"loss": 1.1049,
"step": 400
},
{
"epoch": 2.7739726027397262,
"grad_norm": 0.400390625,
"learning_rate": 0.0001956374630669672,
"loss": 1.1128,
"step": 405
},
{
"epoch": 2.808219178082192,
"grad_norm": 0.478515625,
"learning_rate": 0.00019540160383254107,
"loss": 1.1057,
"step": 410
},
{
"epoch": 2.8424657534246576,
"grad_norm": 0.462890625,
"learning_rate": 0.00019515968533400673,
"loss": 1.1149,
"step": 415
},
{
"epoch": 2.8767123287671232,
"grad_norm": 0.39453125,
"learning_rate": 0.00019491172293638968,
"loss": 1.1132,
"step": 420
},
{
"epoch": 2.910958904109589,
"grad_norm": 0.453125,
"learning_rate": 0.00019465773238858298,
"loss": 1.1059,
"step": 425
},
{
"epoch": 2.9452054794520546,
"grad_norm": 0.43359375,
"learning_rate": 0.00019439772982234697,
"loss": 1.1041,
"step": 430
},
{
"epoch": 2.9794520547945207,
"grad_norm": 0.451171875,
"learning_rate": 0.00019413173175128473,
"loss": 1.1138,
"step": 435
},
{
"epoch": 3.0,
"eval_loss": 2.4846999645233154,
"eval_runtime": 0.5494,
"eval_samples_per_second": 18.201,
"eval_steps_per_second": 1.82,
"step": 438
},
{
"epoch": 3.0136986301369864,
"grad_norm": 0.546875,
"learning_rate": 0.0001938597550697932,
"loss": 1.0911,
"step": 440
},
{
"epoch": 3.047945205479452,
"grad_norm": 0.453125,
"learning_rate": 0.00019358181705199015,
"loss": 1.0876,
"step": 445
},
{
"epoch": 3.0821917808219177,
"grad_norm": 0.400390625,
"learning_rate": 0.00019329793535061723,
"loss": 1.0864,
"step": 450
},
{
"epoch": 3.1164383561643834,
"grad_norm": 0.400390625,
"learning_rate": 0.00019300812799591846,
"loss": 1.0891,
"step": 455
},
{
"epoch": 3.1506849315068495,
"grad_norm": 0.65234375,
"learning_rate": 0.00019271241339449536,
"loss": 1.0752,
"step": 460
},
{
"epoch": 3.184931506849315,
"grad_norm": 0.671875,
"learning_rate": 0.00019241081032813772,
"loss": 1.0834,
"step": 465
},
{
"epoch": 3.219178082191781,
"grad_norm": 0.3828125,
"learning_rate": 0.00019210333795263075,
"loss": 1.0742,
"step": 470
},
{
"epoch": 3.2534246575342465,
"grad_norm": 1.1171875,
"learning_rate": 0.00019179001579653853,
"loss": 1.0925,
"step": 475
},
{
"epoch": 3.287671232876712,
"grad_norm": 0.4765625,
"learning_rate": 0.0001914708637599636,
"loss": 1.0828,
"step": 480
},
{
"epoch": 3.3219178082191783,
"grad_norm": 0.443359375,
"learning_rate": 0.00019114590211328288,
"loss": 1.08,
"step": 485
},
{
"epoch": 3.356164383561644,
"grad_norm": 0.37109375,
"learning_rate": 0.0001908151514958606,
"loss": 1.0891,
"step": 490
},
{
"epoch": 3.3904109589041096,
"grad_norm": 0.58984375,
"learning_rate": 0.00019047863291473717,
"loss": 1.0785,
"step": 495
},
{
"epoch": 3.4246575342465753,
"grad_norm": 0.396484375,
"learning_rate": 0.00019013636774329495,
"loss": 1.0865,
"step": 500
},
{
"epoch": 3.458904109589041,
"grad_norm": 0.671875,
"learning_rate": 0.00018978837771990085,
"loss": 1.0759,
"step": 505
},
{
"epoch": 3.493150684931507,
"grad_norm": 0.396484375,
"learning_rate": 0.0001894346849465257,
"loss": 1.0795,
"step": 510
},
{
"epoch": 3.5273972602739727,
"grad_norm": 0.484375,
"learning_rate": 0.00018907531188734026,
"loss": 1.0868,
"step": 515
},
{
"epoch": 3.5616438356164384,
"grad_norm": 0.55078125,
"learning_rate": 0.00018871028136728874,
"loss": 1.0838,
"step": 520
},
{
"epoch": 3.595890410958904,
"grad_norm": 0.40234375,
"learning_rate": 0.00018833961657063885,
"loss": 1.0748,
"step": 525
},
{
"epoch": 3.6301369863013697,
"grad_norm": 0.59765625,
"learning_rate": 0.0001879633410395095,
"loss": 1.0801,
"step": 530
},
{
"epoch": 3.6643835616438354,
"grad_norm": 0.365234375,
"learning_rate": 0.00018758147867237548,
"loss": 1.0866,
"step": 535
},
{
"epoch": 3.6986301369863015,
"grad_norm": 0.498046875,
"learning_rate": 0.00018719405372254948,
"loss": 1.08,
"step": 540
},
{
"epoch": 3.732876712328767,
"grad_norm": 0.408203125,
"learning_rate": 0.00018680109079664188,
"loss": 1.0704,
"step": 545
},
{
"epoch": 3.767123287671233,
"grad_norm": 0.3671875,
"learning_rate": 0.0001864026148529978,
"loss": 1.0698,
"step": 550
},
{
"epoch": 3.8013698630136985,
"grad_norm": 0.4296875,
"learning_rate": 0.00018599865120011192,
"loss": 1.0801,
"step": 555
},
{
"epoch": 3.8356164383561646,
"grad_norm": 0.3515625,
"learning_rate": 0.00018558922549502107,
"loss": 1.0605,
"step": 560
},
{
"epoch": 3.8698630136986303,
"grad_norm": 0.443359375,
"learning_rate": 0.0001851743637416747,
"loss": 1.0766,
"step": 565
},
{
"epoch": 3.904109589041096,
"grad_norm": 0.345703125,
"learning_rate": 0.00018475409228928312,
"loss": 1.0593,
"step": 570
},
{
"epoch": 3.9383561643835616,
"grad_norm": 0.515625,
"learning_rate": 0.00018432843783064429,
"loss": 1.0751,
"step": 575
},
{
"epoch": 3.9726027397260273,
"grad_norm": 0.458984375,
"learning_rate": 0.00018389742740044813,
"loss": 1.0703,
"step": 580
},
{
"epoch": 4.0,
"eval_loss": 2.492682933807373,
"eval_runtime": 0.5379,
"eval_samples_per_second": 18.592,
"eval_steps_per_second": 1.859,
"step": 584
},
{
"epoch": 4.006849315068493,
"grad_norm": 0.6484375,
"learning_rate": 0.00018346108837355972,
"loss": 1.0779,
"step": 585
},
{
"epoch": 4.041095890410959,
"grad_norm": 0.4140625,
"learning_rate": 0.00018301944846328049,
"loss": 1.0393,
"step": 590
},
{
"epoch": 4.075342465753424,
"grad_norm": 0.390625,
"learning_rate": 0.0001825725357195881,
"loss": 1.0573,
"step": 595
},
{
"epoch": 4.109589041095891,
"grad_norm": 0.412109375,
"learning_rate": 0.00018212037852735486,
"loss": 1.0498,
"step": 600
},
{
"epoch": 4.1438356164383565,
"grad_norm": 0.353515625,
"learning_rate": 0.0001816630056045451,
"loss": 1.0507,
"step": 605
},
{
"epoch": 4.178082191780822,
"grad_norm": 0.384765625,
"learning_rate": 0.0001812004460003909,
"loss": 1.0398,
"step": 610
},
{
"epoch": 4.212328767123288,
"grad_norm": 0.45703125,
"learning_rate": 0.00018073272909354727,
"loss": 1.0603,
"step": 615
},
{
"epoch": 4.2465753424657535,
"grad_norm": 0.5703125,
"learning_rate": 0.0001802598845902262,
"loss": 1.0383,
"step": 620
},
{
"epoch": 4.280821917808219,
"grad_norm": 0.40625,
"learning_rate": 0.00017978194252230985,
"loss": 1.0511,
"step": 625
},
{
"epoch": 4.315068493150685,
"grad_norm": 0.39453125,
"learning_rate": 0.00017929893324544332,
"loss": 1.0423,
"step": 630
},
{
"epoch": 4.3493150684931505,
"grad_norm": 0.65234375,
"learning_rate": 0.0001788108874371063,
"loss": 1.0549,
"step": 635
},
{
"epoch": 4.383561643835616,
"grad_norm": 0.388671875,
"learning_rate": 0.00017831783609466504,
"loss": 1.0481,
"step": 640
},
{
"epoch": 4.417808219178082,
"grad_norm": 0.373046875,
"learning_rate": 0.00017781981053340337,
"loss": 1.0571,
"step": 645
},
{
"epoch": 4.4520547945205475,
"grad_norm": 0.384765625,
"learning_rate": 0.00017731684238453385,
"loss": 1.0452,
"step": 650
},
{
"epoch": 4.486301369863014,
"grad_norm": 0.427734375,
"learning_rate": 0.0001768089635931887,
"loss": 1.0548,
"step": 655
},
{
"epoch": 4.52054794520548,
"grad_norm": 0.671875,
"learning_rate": 0.00017629620641639103,
"loss": 1.0489,
"step": 660
},
{
"epoch": 4.554794520547945,
"grad_norm": 0.478515625,
"learning_rate": 0.00017577860342100579,
"loss": 1.0554,
"step": 665
},
{
"epoch": 4.589041095890411,
"grad_norm": 0.74609375,
"learning_rate": 0.0001752561874816717,
"loss": 1.0573,
"step": 670
},
{
"epoch": 4.623287671232877,
"grad_norm": 0.63671875,
"learning_rate": 0.00017472899177871297,
"loss": 1.0491,
"step": 675
},
{
"epoch": 4.657534246575342,
"grad_norm": 0.365234375,
"learning_rate": 0.00017419704979603214,
"loss": 1.0613,
"step": 680
},
{
"epoch": 4.691780821917808,
"grad_norm": 0.431640625,
"learning_rate": 0.00017366039531898326,
"loss": 1.0561,
"step": 685
},
{
"epoch": 4.726027397260274,
"grad_norm": 0.373046875,
"learning_rate": 0.00017311906243222614,
"loss": 1.0588,
"step": 690
},
{
"epoch": 4.760273972602739,
"grad_norm": 0.361328125,
"learning_rate": 0.0001725730855175615,
"loss": 1.061,
"step": 695
},
{
"epoch": 4.794520547945205,
"grad_norm": 0.419921875,
"learning_rate": 0.00017202249925174723,
"loss": 1.0454,
"step": 700
},
{
"epoch": 4.828767123287671,
"grad_norm": 0.484375,
"learning_rate": 0.00017146733860429612,
"loss": 1.0584,
"step": 705
},
{
"epoch": 4.863013698630137,
"grad_norm": 0.50390625,
"learning_rate": 0.0001709076388352546,
"loss": 1.0478,
"step": 710
},
{
"epoch": 4.897260273972603,
"grad_norm": 0.337890625,
"learning_rate": 0.00017034343549296346,
"loss": 1.068,
"step": 715
},
{
"epoch": 4.931506849315069,
"grad_norm": 0.5234375,
"learning_rate": 0.00016977476441179992,
"loss": 1.0435,
"step": 720
},
{
"epoch": 4.965753424657534,
"grad_norm": 0.50390625,
"learning_rate": 0.0001692016617099018,
"loss": 1.0474,
"step": 725
},
{
"epoch": 5.0,
"grad_norm": 0.453125,
"learning_rate": 0.0001686241637868734,
"loss": 1.0423,
"step": 730
},
{
"epoch": 5.0,
"eval_loss": 2.5079874992370605,
"eval_runtime": 0.5486,
"eval_samples_per_second": 18.227,
"eval_steps_per_second": 1.823,
"step": 730
},
{
"epoch": 5.034246575342466,
"grad_norm": 0.38671875,
"learning_rate": 0.0001680423073214737,
"loss": 1.035,
"step": 735
},
{
"epoch": 5.068493150684931,
"grad_norm": 0.376953125,
"learning_rate": 0.00016745612926928694,
"loss": 1.0372,
"step": 740
},
{
"epoch": 5.102739726027397,
"grad_norm": 0.43359375,
"learning_rate": 0.0001668656668603751,
"loss": 1.0247,
"step": 745
},
{
"epoch": 5.136986301369863,
"grad_norm": 0.36328125,
"learning_rate": 0.00016627095759691362,
"loss": 1.0201,
"step": 750
},
{
"epoch": 5.171232876712328,
"grad_norm": 0.44140625,
"learning_rate": 0.0001656720392508094,
"loss": 1.0256,
"step": 755
},
{
"epoch": 5.205479452054795,
"grad_norm": 0.376953125,
"learning_rate": 0.00016506894986130171,
"loss": 1.024,
"step": 760
},
{
"epoch": 5.239726027397261,
"grad_norm": 0.484375,
"learning_rate": 0.00016446172773254629,
"loss": 1.0239,
"step": 765
},
{
"epoch": 5.273972602739726,
"grad_norm": 0.6015625,
"learning_rate": 0.00016385041143118255,
"loss": 1.0331,
"step": 770
},
{
"epoch": 5.308219178082192,
"grad_norm": 0.515625,
"learning_rate": 0.000163235039783884,
"loss": 1.0368,
"step": 775
},
{
"epoch": 5.342465753424658,
"grad_norm": 0.5859375,
"learning_rate": 0.0001626156518748922,
"loss": 1.0276,
"step": 780
},
{
"epoch": 5.376712328767123,
"grad_norm": 0.546875,
"learning_rate": 0.00016199228704353455,
"loss": 1.0396,
"step": 785
},
{
"epoch": 5.410958904109589,
"grad_norm": 0.98046875,
"learning_rate": 0.00016136498488172568,
"loss": 1.0279,
"step": 790
},
{
"epoch": 5.445205479452055,
"grad_norm": 0.392578125,
"learning_rate": 0.0001607337852314527,
"loss": 1.0378,
"step": 795
},
{
"epoch": 5.47945205479452,
"grad_norm": 0.400390625,
"learning_rate": 0.00016009872818224485,
"loss": 1.0398,
"step": 800
},
{
"epoch": 5.513698630136986,
"grad_norm": 0.404296875,
"learning_rate": 0.00015945985406862721,
"loss": 1.031,
"step": 805
},
{
"epoch": 5.5479452054794525,
"grad_norm": 0.431640625,
"learning_rate": 0.00015881720346755905,
"loss": 1.0248,
"step": 810
},
{
"epoch": 5.582191780821918,
"grad_norm": 0.40234375,
"learning_rate": 0.00015817081719585643,
"loss": 1.023,
"step": 815
},
{
"epoch": 5.616438356164384,
"grad_norm": 0.44140625,
"learning_rate": 0.00015752073630759998,
"loss": 1.0427,
"step": 820
},
{
"epoch": 5.6506849315068495,
"grad_norm": 0.431640625,
"learning_rate": 0.00015686700209152738,
"loss": 1.0256,
"step": 825
},
{
"epoch": 5.684931506849315,
"grad_norm": 0.359375,
"learning_rate": 0.00015620965606841098,
"loss": 1.0225,
"step": 830
},
{
"epoch": 5.719178082191781,
"grad_norm": 0.40625,
"learning_rate": 0.0001555487399884206,
"loss": 1.0237,
"step": 835
},
{
"epoch": 5.7534246575342465,
"grad_norm": 0.455078125,
"learning_rate": 0.00015488429582847192,
"loss": 1.02,
"step": 840
},
{
"epoch": 5.787671232876712,
"grad_norm": 0.408203125,
"learning_rate": 0.0001542163657895605,
"loss": 1.0223,
"step": 845
},
{
"epoch": 5.821917808219178,
"grad_norm": 0.51171875,
"learning_rate": 0.00015354499229408114,
"loss": 1.0246,
"step": 850
},
{
"epoch": 5.8561643835616435,
"grad_norm": 0.54296875,
"learning_rate": 0.0001528702179831338,
"loss": 1.0216,
"step": 855
},
{
"epoch": 5.890410958904109,
"grad_norm": 0.546875,
"learning_rate": 0.00015219208571381525,
"loss": 1.028,
"step": 860
},
{
"epoch": 5.924657534246576,
"grad_norm": 0.46875,
"learning_rate": 0.00015151063855649698,
"loss": 1.0404,
"step": 865
},
{
"epoch": 5.958904109589041,
"grad_norm": 0.369140625,
"learning_rate": 0.00015082591979208976,
"loss": 1.0311,
"step": 870
},
{
"epoch": 5.993150684931507,
"grad_norm": 0.63671875,
"learning_rate": 0.00015013797290929466,
"loss": 1.0322,
"step": 875
},
{
"epoch": 6.0,
"eval_loss": 2.5202295780181885,
"eval_runtime": 0.5425,
"eval_samples_per_second": 18.433,
"eval_steps_per_second": 1.843,
"step": 876
},
{
"epoch": 6.027397260273973,
"grad_norm": 0.515625,
"learning_rate": 0.00014944684160184108,
"loss": 1.0176,
"step": 880
},
{
"epoch": 6.061643835616438,
"grad_norm": 0.412109375,
"learning_rate": 0.00014875256976571135,
"loss": 1.0053,
"step": 885
},
{
"epoch": 6.095890410958904,
"grad_norm": 0.3984375,
"learning_rate": 0.00014805520149635307,
"loss": 0.9952,
"step": 890
},
{
"epoch": 6.13013698630137,
"grad_norm": 0.3828125,
"learning_rate": 0.00014735478108587828,
"loss": 1.0034,
"step": 895
},
{
"epoch": 6.164383561643835,
"grad_norm": 0.431640625,
"learning_rate": 0.00014665135302025035,
"loss": 1.0069,
"step": 900
},
{
"epoch": 6.198630136986301,
"grad_norm": 0.404296875,
"learning_rate": 0.00014594496197645852,
"loss": 1.0014,
"step": 905
},
{
"epoch": 6.232876712328767,
"grad_norm": 0.392578125,
"learning_rate": 0.0001452356528196804,
"loss": 1.0081,
"step": 910
},
{
"epoch": 6.267123287671233,
"grad_norm": 0.37890625,
"learning_rate": 0.00014452347060043237,
"loss": 1.0132,
"step": 915
},
{
"epoch": 6.301369863013699,
"grad_norm": 0.4375,
"learning_rate": 0.00014380846055170828,
"loss": 1.0091,
"step": 920
},
{
"epoch": 6.335616438356165,
"grad_norm": 0.5546875,
"learning_rate": 0.00014309066808610655,
"loss": 1.0125,
"step": 925
},
{
"epoch": 6.36986301369863,
"grad_norm": 0.498046875,
"learning_rate": 0.0001423701387929459,
"loss": 1.012,
"step": 930
},
{
"epoch": 6.404109589041096,
"grad_norm": 0.390625,
"learning_rate": 0.00014164691843536982,
"loss": 1.0092,
"step": 935
},
{
"epoch": 6.438356164383562,
"grad_norm": 0.42578125,
"learning_rate": 0.00014092105294744,
"loss": 1.0142,
"step": 940
},
{
"epoch": 6.472602739726027,
"grad_norm": 0.423828125,
"learning_rate": 0.00014019258843121893,
"loss": 1.0131,
"step": 945
},
{
"epoch": 6.506849315068493,
"grad_norm": 0.419921875,
"learning_rate": 0.0001394615711538417,
"loss": 1.0083,
"step": 950
},
{
"epoch": 6.541095890410959,
"grad_norm": 0.40625,
"learning_rate": 0.00013872804754457759,
"loss": 1.0118,
"step": 955
},
{
"epoch": 6.575342465753424,
"grad_norm": 0.408203125,
"learning_rate": 0.00013799206419188103,
"loss": 1.0183,
"step": 960
},
{
"epoch": 6.609589041095891,
"grad_norm": 0.40234375,
"learning_rate": 0.00013725366784043288,
"loss": 1.0103,
"step": 965
},
{
"epoch": 6.6438356164383565,
"grad_norm": 0.40234375,
"learning_rate": 0.00013651290538817113,
"loss": 1.0134,
"step": 970
},
{
"epoch": 6.678082191780822,
"grad_norm": 0.423828125,
"learning_rate": 0.0001357698238833126,
"loss": 1.0199,
"step": 975
},
{
"epoch": 6.712328767123288,
"grad_norm": 0.3984375,
"learning_rate": 0.00013502447052136455,
"loss": 1.0042,
"step": 980
},
{
"epoch": 6.7465753424657535,
"grad_norm": 0.38671875,
"learning_rate": 0.00013427689264212738,
"loss": 1.0171,
"step": 985
},
{
"epoch": 6.780821917808219,
"grad_norm": 0.423828125,
"learning_rate": 0.00013352713772668765,
"loss": 1.0065,
"step": 990
},
{
"epoch": 6.815068493150685,
"grad_norm": 0.408203125,
"learning_rate": 0.0001327752533944025,
"loss": 1.01,
"step": 995
},
{
"epoch": 6.8493150684931505,
"grad_norm": 0.515625,
"learning_rate": 0.00013202128739987532,
"loss": 1.0133,
"step": 1000
},
{
"epoch": 6.883561643835616,
"grad_norm": 0.447265625,
"learning_rate": 0.00013126528762992247,
"loss": 1.0171,
"step": 1005
},
{
"epoch": 6.917808219178082,
"grad_norm": 0.39453125,
"learning_rate": 0.0001305073021005321,
"loss": 1.011,
"step": 1010
},
{
"epoch": 6.9520547945205475,
"grad_norm": 0.451171875,
"learning_rate": 0.0001297473789538142,
"loss": 1.0125,
"step": 1015
},
{
"epoch": 6.986301369863014,
"grad_norm": 0.3984375,
"learning_rate": 0.00012898556645494325,
"loss": 1.0113,
"step": 1020
},
{
"epoch": 7.0,
"eval_loss": 2.538480758666992,
"eval_runtime": 0.546,
"eval_samples_per_second": 18.313,
"eval_steps_per_second": 1.831,
"step": 1022
},
{
"epoch": 7.02054794520548,
"grad_norm": 0.400390625,
"learning_rate": 0.0001282219129890925,
"loss": 0.9999,
"step": 1025
},
{
"epoch": 7.054794520547945,
"grad_norm": 0.40234375,
"learning_rate": 0.00012745646705836097,
"loss": 0.9939,
"step": 1030
},
{
"epoch": 7.089041095890411,
"grad_norm": 0.54296875,
"learning_rate": 0.0001266892772786929,
"loss": 0.9812,
"step": 1035
},
{
"epoch": 7.123287671232877,
"grad_norm": 0.3984375,
"learning_rate": 0.0001259203923767901,
"loss": 0.9884,
"step": 1040
},
{
"epoch": 7.157534246575342,
"grad_norm": 0.38671875,
"learning_rate": 0.00012514986118701695,
"loss": 0.988,
"step": 1045
},
{
"epoch": 7.191780821917808,
"grad_norm": 0.416015625,
"learning_rate": 0.00012437773264829897,
"loss": 0.9939,
"step": 1050
},
{
"epoch": 7.226027397260274,
"grad_norm": 0.416015625,
"learning_rate": 0.00012360405580101448,
"loss": 0.9979,
"step": 1055
},
{
"epoch": 7.260273972602739,
"grad_norm": 0.443359375,
"learning_rate": 0.00012282887978387976,
"loss": 1.0047,
"step": 1060
},
{
"epoch": 7.294520547945205,
"grad_norm": 0.458984375,
"learning_rate": 0.00012205225383082843,
"loss": 0.9971,
"step": 1065
},
{
"epoch": 7.328767123287671,
"grad_norm": 0.439453125,
"learning_rate": 0.000121274227267884,
"loss": 0.9918,
"step": 1070
},
{
"epoch": 7.363013698630137,
"grad_norm": 0.421875,
"learning_rate": 0.00012049484951002739,
"loss": 1.0011,
"step": 1075
},
{
"epoch": 7.397260273972603,
"grad_norm": 0.416015625,
"learning_rate": 0.00011971417005805818,
"loss": 0.9909,
"step": 1080
},
{
"epoch": 7.431506849315069,
"grad_norm": 0.376953125,
"learning_rate": 0.00011893223849545084,
"loss": 0.9974,
"step": 1085
},
{
"epoch": 7.465753424657534,
"grad_norm": 0.408203125,
"learning_rate": 0.00011814910448520536,
"loss": 0.9934,
"step": 1090
},
{
"epoch": 7.5,
"grad_norm": 0.404296875,
"learning_rate": 0.00011736481776669306,
"loss": 0.9965,
"step": 1095
},
{
"epoch": 7.534246575342466,
"grad_norm": 0.396484375,
"learning_rate": 0.00011657942815249754,
"loss": 0.9936,
"step": 1100
},
{
"epoch": 7.568493150684931,
"grad_norm": 0.38671875,
"learning_rate": 0.00011579298552525084,
"loss": 0.9919,
"step": 1105
},
{
"epoch": 7.602739726027397,
"grad_norm": 0.408203125,
"learning_rate": 0.00011500553983446527,
"loss": 0.9956,
"step": 1110
},
{
"epoch": 7.636986301369863,
"grad_norm": 0.39453125,
"learning_rate": 0.00011421714109336097,
"loss": 0.9989,
"step": 1115
},
{
"epoch": 7.671232876712329,
"grad_norm": 0.46875,
"learning_rate": 0.00011342783937568926,
"loss": 1.0018,
"step": 1120
},
{
"epoch": 7.705479452054795,
"grad_norm": 0.46484375,
"learning_rate": 0.00011263768481255264,
"loss": 0.993,
"step": 1125
},
{
"epoch": 7.739726027397261,
"grad_norm": 0.419921875,
"learning_rate": 0.00011184672758922034,
"loss": 1.0063,
"step": 1130
},
{
"epoch": 7.773972602739726,
"grad_norm": 0.44140625,
"learning_rate": 0.00011105501794194131,
"loss": 0.9936,
"step": 1135
},
{
"epoch": 7.808219178082192,
"grad_norm": 0.427734375,
"learning_rate": 0.00011026260615475333,
"loss": 1.0078,
"step": 1140
},
{
"epoch": 7.842465753424658,
"grad_norm": 0.38671875,
"learning_rate": 0.00010946954255628928,
"loss": 0.9997,
"step": 1145
},
{
"epoch": 7.876712328767123,
"grad_norm": 0.478515625,
"learning_rate": 0.00010867587751658079,
"loss": 0.9902,
"step": 1150
},
{
"epoch": 7.910958904109589,
"grad_norm": 0.38671875,
"learning_rate": 0.00010788166144385888,
"loss": 0.9905,
"step": 1155
},
{
"epoch": 7.945205479452055,
"grad_norm": 0.431640625,
"learning_rate": 0.0001070869447813525,
"loss": 0.9834,
"step": 1160
},
{
"epoch": 7.97945205479452,
"grad_norm": 0.392578125,
"learning_rate": 0.0001062917780040847,
"loss": 0.9857,
"step": 1165
},
{
"epoch": 8.0,
"eval_loss": 2.55222225189209,
"eval_runtime": 0.5458,
"eval_samples_per_second": 18.321,
"eval_steps_per_second": 1.832,
"step": 1168
},
{
"epoch": 8.013698630136986,
"grad_norm": 0.412109375,
"learning_rate": 0.0001054962116156667,
"loss": 0.9837,
"step": 1170
},
{
"epoch": 8.047945205479452,
"grad_norm": 0.41796875,
"learning_rate": 0.00010470029614509041,
"loss": 0.9743,
"step": 1175
},
{
"epoch": 8.082191780821917,
"grad_norm": 0.423828125,
"learning_rate": 0.00010390408214351892,
"loss": 0.9731,
"step": 1180
},
{
"epoch": 8.116438356164384,
"grad_norm": 0.427734375,
"learning_rate": 0.0001031076201810762,
"loss": 0.9772,
"step": 1185
},
{
"epoch": 8.150684931506849,
"grad_norm": 0.4296875,
"learning_rate": 0.00010231096084363483,
"loss": 0.9671,
"step": 1190
},
{
"epoch": 8.184931506849315,
"grad_norm": 0.40234375,
"learning_rate": 0.00010151415472960342,
"loss": 0.9934,
"step": 1195
},
{
"epoch": 8.219178082191782,
"grad_norm": 0.423828125,
"learning_rate": 0.00010071725244671282,
"loss": 0.9819,
"step": 1200
},
{
"epoch": 8.253424657534246,
"grad_norm": 0.3984375,
"learning_rate": 9.992030460880181e-05,
"loss": 0.9712,
"step": 1205
},
{
"epoch": 8.287671232876713,
"grad_norm": 0.40234375,
"learning_rate": 9.91233618326026e-05,
"loss": 0.9846,
"step": 1210
},
{
"epoch": 8.321917808219178,
"grad_norm": 0.4296875,
"learning_rate": 9.83264747345259e-05,
"loss": 0.98,
"step": 1215
},
{
"epoch": 8.356164383561644,
"grad_norm": 0.423828125,
"learning_rate": 9.752969392744606e-05,
"loss": 0.9835,
"step": 1220
},
{
"epoch": 8.39041095890411,
"grad_norm": 0.44921875,
"learning_rate": 9.673307001748661e-05,
"loss": 0.9739,
"step": 1225
},
{
"epoch": 8.424657534246576,
"grad_norm": 0.435546875,
"learning_rate": 9.593665360080599e-05,
"loss": 0.9735,
"step": 1230
},
{
"epoch": 8.45890410958904,
"grad_norm": 0.42578125,
"learning_rate": 9.514049526038418e-05,
"loss": 0.9816,
"step": 1235
},
{
"epoch": 8.493150684931507,
"grad_norm": 0.44140625,
"learning_rate": 9.43446455628097e-05,
"loss": 0.9813,
"step": 1240
},
{
"epoch": 8.527397260273972,
"grad_norm": 0.455078125,
"learning_rate": 9.354915505506839e-05,
"loss": 0.9802,
"step": 1245
},
{
"epoch": 8.561643835616438,
"grad_norm": 0.43359375,
"learning_rate": 9.27540742613326e-05,
"loss": 0.9836,
"step": 1250
},
{
"epoch": 8.595890410958905,
"grad_norm": 0.427734375,
"learning_rate": 9.195945367975256e-05,
"loss": 0.974,
"step": 1255
},
{
"epoch": 8.63013698630137,
"grad_norm": 0.453125,
"learning_rate": 9.116534377924883e-05,
"loss": 0.9819,
"step": 1260
},
{
"epoch": 8.664383561643836,
"grad_norm": 0.421875,
"learning_rate": 9.037179499630703e-05,
"loss": 0.9756,
"step": 1265
},
{
"epoch": 8.698630136986301,
"grad_norm": 0.4375,
"learning_rate": 8.957885773177438e-05,
"loss": 0.9929,
"step": 1270
},
{
"epoch": 8.732876712328768,
"grad_norm": 0.62109375,
"learning_rate": 8.878658234765858e-05,
"loss": 0.9788,
"step": 1275
},
{
"epoch": 8.767123287671232,
"grad_norm": 0.458984375,
"learning_rate": 8.799501916392912e-05,
"loss": 0.9897,
"step": 1280
},
{
"epoch": 8.801369863013699,
"grad_norm": 0.40234375,
"learning_rate": 8.720421845532151e-05,
"loss": 0.9844,
"step": 1285
},
{
"epoch": 8.835616438356164,
"grad_norm": 0.42578125,
"learning_rate": 8.641423044814374e-05,
"loss": 0.9809,
"step": 1290
},
{
"epoch": 8.86986301369863,
"grad_norm": 0.42578125,
"learning_rate": 8.562510531708677e-05,
"loss": 0.9889,
"step": 1295
},
{
"epoch": 8.904109589041095,
"grad_norm": 0.40625,
"learning_rate": 8.48368931820373e-05,
"loss": 0.9838,
"step": 1300
},
{
"epoch": 8.938356164383562,
"grad_norm": 0.41796875,
"learning_rate": 8.404964410489485e-05,
"loss": 0.9877,
"step": 1305
},
{
"epoch": 8.972602739726028,
"grad_norm": 0.427734375,
"learning_rate": 8.32634080863919e-05,
"loss": 0.9865,
"step": 1310
},
{
"epoch": 9.0,
"eval_loss": 2.5657296180725098,
"eval_runtime": 0.5493,
"eval_samples_per_second": 18.206,
"eval_steps_per_second": 1.821,
"step": 1314
},
{
"epoch": 9.006849315068493,
"grad_norm": 0.41796875,
"learning_rate": 8.247823506291844e-05,
"loss": 0.9785,
"step": 1315
},
{
"epoch": 9.04109589041096,
"grad_norm": 0.41796875,
"learning_rate": 8.169417490335007e-05,
"loss": 0.9678,
"step": 1320
},
{
"epoch": 9.075342465753424,
"grad_norm": 0.4609375,
"learning_rate": 8.091127740588094e-05,
"loss": 0.9555,
"step": 1325
},
{
"epoch": 9.10958904109589,
"grad_norm": 0.431640625,
"learning_rate": 8.012959229486061e-05,
"loss": 0.9599,
"step": 1330
},
{
"epoch": 9.143835616438356,
"grad_norm": 0.46875,
"learning_rate": 7.934916921763628e-05,
"loss": 0.9644,
"step": 1335
},
{
"epoch": 9.178082191780822,
"grad_norm": 0.427734375,
"learning_rate": 7.857005774139907e-05,
"loss": 0.964,
"step": 1340
},
{
"epoch": 9.212328767123287,
"grad_norm": 0.447265625,
"learning_rate": 7.779230735003628e-05,
"loss": 0.9671,
"step": 1345
},
{
"epoch": 9.246575342465754,
"grad_norm": 0.396484375,
"learning_rate": 7.701596744098818e-05,
"loss": 0.962,
"step": 1350
},
{
"epoch": 9.280821917808218,
"grad_norm": 0.4140625,
"learning_rate": 7.624108732211081e-05,
"loss": 0.9547,
"step": 1355
},
{
"epoch": 9.315068493150685,
"grad_norm": 0.4140625,
"learning_rate": 7.54677162085442e-05,
"loss": 0.9729,
"step": 1360
},
{
"epoch": 9.349315068493151,
"grad_norm": 0.40625,
"learning_rate": 7.469590321958662e-05,
"loss": 0.9658,
"step": 1365
},
{
"epoch": 9.383561643835616,
"grad_norm": 0.43359375,
"learning_rate": 7.392569737557474e-05,
"loss": 0.9712,
"step": 1370
},
{
"epoch": 9.417808219178083,
"grad_norm": 0.455078125,
"learning_rate": 7.31571475947703e-05,
"loss": 0.9658,
"step": 1375
},
{
"epoch": 9.452054794520548,
"grad_norm": 0.416015625,
"learning_rate": 7.239030269025311e-05,
"loss": 0.9634,
"step": 1380
},
{
"epoch": 9.486301369863014,
"grad_norm": 0.390625,
"learning_rate": 7.162521136682085e-05,
"loss": 0.9794,
"step": 1385
},
{
"epoch": 9.520547945205479,
"grad_norm": 0.4375,
"learning_rate": 7.08619222178954e-05,
"loss": 0.9713,
"step": 1390
},
{
"epoch": 9.554794520547945,
"grad_norm": 0.412109375,
"learning_rate": 7.010048372243698e-05,
"loss": 0.974,
"step": 1395
},
{
"epoch": 9.58904109589041,
"grad_norm": 0.439453125,
"learning_rate": 6.934094424186459e-05,
"loss": 0.9682,
"step": 1400
},
{
"epoch": 9.623287671232877,
"grad_norm": 0.416015625,
"learning_rate": 6.858335201698485e-05,
"loss": 0.9787,
"step": 1405
},
{
"epoch": 9.657534246575342,
"grad_norm": 0.4453125,
"learning_rate": 6.782775516492771e-05,
"loss": 0.9623,
"step": 1410
},
{
"epoch": 9.691780821917808,
"grad_norm": 0.43359375,
"learning_rate": 6.70742016760907e-05,
"loss": 0.975,
"step": 1415
},
{
"epoch": 9.726027397260275,
"grad_norm": 0.44921875,
"learning_rate": 6.632273941109064e-05,
"loss": 0.9571,
"step": 1420
},
{
"epoch": 9.76027397260274,
"grad_norm": 0.400390625,
"learning_rate": 6.5573416097724e-05,
"loss": 0.9782,
"step": 1425
},
{
"epoch": 9.794520547945206,
"grad_norm": 0.451171875,
"learning_rate": 6.482627932793553e-05,
"loss": 0.9863,
"step": 1430
},
{
"epoch": 9.82876712328767,
"grad_norm": 0.42578125,
"learning_rate": 6.408137655479554e-05,
"loss": 0.985,
"step": 1435
},
{
"epoch": 9.863013698630137,
"grad_norm": 0.45703125,
"learning_rate": 6.333875508948593e-05,
"loss": 0.9727,
"step": 1440
},
{
"epoch": 9.897260273972602,
"grad_norm": 0.41796875,
"learning_rate": 6.259846209829551e-05,
"loss": 0.9875,
"step": 1445
},
{
"epoch": 9.931506849315069,
"grad_norm": 0.453125,
"learning_rate": 6.186054459962399e-05,
"loss": 0.9715,
"step": 1450
},
{
"epoch": 9.965753424657533,
"grad_norm": 0.451171875,
"learning_rate": 6.112504946099604e-05,
"loss": 0.969,
"step": 1455
},
{
"epoch": 10.0,
"grad_norm": 0.416015625,
"learning_rate": 6.039202339608432e-05,
"loss": 0.9691,
"step": 1460
},
{
"epoch": 10.0,
"eval_loss": 2.5773940086364746,
"eval_runtime": 0.5541,
"eval_samples_per_second": 18.049,
"eval_steps_per_second": 1.805,
"step": 1460
},
{
"epoch": 10.034246575342467,
"grad_norm": 0.404296875,
"learning_rate": 5.966151296174268e-05,
"loss": 0.9569,
"step": 1465
},
{
"epoch": 10.068493150684931,
"grad_norm": 0.447265625,
"learning_rate": 5.8933564555049105e-05,
"loss": 0.9608,
"step": 1470
},
{
"epoch": 10.102739726027398,
"grad_norm": 0.439453125,
"learning_rate": 5.820822441035899e-05,
"loss": 0.9571,
"step": 1475
},
{
"epoch": 10.136986301369863,
"grad_norm": 0.408203125,
"learning_rate": 5.7485538596368496e-05,
"loss": 0.9585,
"step": 1480
},
{
"epoch": 10.17123287671233,
"grad_norm": 0.41015625,
"learning_rate": 5.6765553013188766e-05,
"loss": 0.956,
"step": 1485
},
{
"epoch": 10.205479452054794,
"grad_norm": 0.396484375,
"learning_rate": 5.6048313389430484e-05,
"loss": 0.9522,
"step": 1490
},
{
"epoch": 10.23972602739726,
"grad_norm": 0.40625,
"learning_rate": 5.533386527929962e-05,
"loss": 0.9652,
"step": 1495
},
{
"epoch": 10.273972602739725,
"grad_norm": 0.4296875,
"learning_rate": 5.462225405970401e-05,
"loss": 0.9611,
"step": 1500
},
{
"epoch": 10.308219178082192,
"grad_norm": 0.435546875,
"learning_rate": 5.391352492737157e-05,
"loss": 0.956,
"step": 1505
},
{
"epoch": 10.342465753424657,
"grad_norm": 0.412109375,
"learning_rate": 5.3207722895979406e-05,
"loss": 0.9602,
"step": 1510
},
{
"epoch": 10.376712328767123,
"grad_norm": 0.41015625,
"learning_rate": 5.2504892793295e-05,
"loss": 0.9607,
"step": 1515
},
{
"epoch": 10.41095890410959,
"grad_norm": 0.5,
"learning_rate": 5.1805079258329056e-05,
"loss": 0.9658,
"step": 1520
},
{
"epoch": 10.445205479452055,
"grad_norm": 0.5078125,
"learning_rate": 5.110832673850039e-05,
"loss": 0.9648,
"step": 1525
},
{
"epoch": 10.479452054794521,
"grad_norm": 0.451171875,
"learning_rate": 5.041467948681269e-05,
"loss": 0.9732,
"step": 1530
},
{
"epoch": 10.513698630136986,
"grad_norm": 0.439453125,
"learning_rate": 4.9724181559044234e-05,
"loss": 0.9681,
"step": 1535
},
{
"epoch": 10.547945205479452,
"grad_norm": 0.404296875,
"learning_rate": 4.903687681094942e-05,
"loss": 0.9639,
"step": 1540
},
{
"epoch": 10.582191780821917,
"grad_norm": 0.412109375,
"learning_rate": 4.835280889547351e-05,
"loss": 0.9579,
"step": 1545
},
{
"epoch": 10.616438356164384,
"grad_norm": 0.41015625,
"learning_rate": 4.767202125998005e-05,
"loss": 0.9553,
"step": 1550
},
{
"epoch": 10.650684931506849,
"grad_norm": 0.41796875,
"learning_rate": 4.699455714349152e-05,
"loss": 0.9512,
"step": 1555
},
{
"epoch": 10.684931506849315,
"grad_norm": 0.416015625,
"learning_rate": 4.6320459573942856e-05,
"loss": 0.968,
"step": 1560
},
{
"epoch": 10.719178082191782,
"grad_norm": 0.416015625,
"learning_rate": 4.564977136544873e-05,
"loss": 0.9617,
"step": 1565
},
{
"epoch": 10.753424657534246,
"grad_norm": 0.421875,
"learning_rate": 4.49825351155843e-05,
"loss": 0.9749,
"step": 1570
},
{
"epoch": 10.787671232876713,
"grad_norm": 0.44140625,
"learning_rate": 4.431879320267972e-05,
"loss": 0.9662,
"step": 1575
},
{
"epoch": 10.821917808219178,
"grad_norm": 0.412109375,
"learning_rate": 4.3658587783128425e-05,
"loss": 0.9661,
"step": 1580
},
{
"epoch": 10.856164383561644,
"grad_norm": 0.4609375,
"learning_rate": 4.300196078870982e-05,
"loss": 0.9571,
"step": 1585
},
{
"epoch": 10.89041095890411,
"grad_norm": 0.41015625,
"learning_rate": 4.2348953923925916e-05,
"loss": 0.9657,
"step": 1590
},
{
"epoch": 10.924657534246576,
"grad_norm": 0.43359375,
"learning_rate": 4.16996086633526e-05,
"loss": 0.9587,
"step": 1595
},
{
"epoch": 10.95890410958904,
"grad_norm": 0.390625,
"learning_rate": 4.105396624900538e-05,
"loss": 0.9548,
"step": 1600
},
{
"epoch": 10.993150684931507,
"grad_norm": 0.427734375,
"learning_rate": 4.041206768772022e-05,
"loss": 0.952,
"step": 1605
},
{
"epoch": 11.0,
"eval_loss": 2.5889077186584473,
"eval_runtime": 0.5538,
"eval_samples_per_second": 18.057,
"eval_steps_per_second": 1.806,
"step": 1606
},
{
"epoch": 11.027397260273972,
"grad_norm": 0.416015625,
"learning_rate": 3.977395374854871e-05,
"loss": 0.9434,
"step": 1610
},
{
"epoch": 11.061643835616438,
"grad_norm": 0.419921875,
"learning_rate": 3.913966496016891e-05,
"loss": 0.9579,
"step": 1615
},
{
"epoch": 11.095890410958905,
"grad_norm": 0.453125,
"learning_rate": 3.850924160831115e-05,
"loss": 0.9524,
"step": 1620
},
{
"epoch": 11.13013698630137,
"grad_norm": 0.42578125,
"learning_rate": 3.788272373319955e-05,
"loss": 0.9618,
"step": 1625
},
{
"epoch": 11.164383561643836,
"grad_norm": 0.40625,
"learning_rate": 3.726015112700859e-05,
"loss": 0.9503,
"step": 1630
},
{
"epoch": 11.198630136986301,
"grad_norm": 0.41796875,
"learning_rate": 3.6641563331336125e-05,
"loss": 0.9475,
"step": 1635
},
{
"epoch": 11.232876712328768,
"grad_norm": 0.41796875,
"learning_rate": 3.6026999634691725e-05,
"loss": 0.945,
"step": 1640
},
{
"epoch": 11.267123287671232,
"grad_norm": 0.40234375,
"learning_rate": 3.541649907000147e-05,
"loss": 0.9528,
"step": 1645
},
{
"epoch": 11.301369863013699,
"grad_norm": 0.423828125,
"learning_rate": 3.4810100412128747e-05,
"loss": 0.946,
"step": 1650
},
{
"epoch": 11.335616438356164,
"grad_norm": 0.404296875,
"learning_rate": 3.42078421754117e-05,
"loss": 0.9535,
"step": 1655
},
{
"epoch": 11.36986301369863,
"grad_norm": 0.41015625,
"learning_rate": 3.360976261121684e-05,
"loss": 0.962,
"step": 1660
},
{
"epoch": 11.404109589041095,
"grad_norm": 0.412109375,
"learning_rate": 3.3015899705509734e-05,
"loss": 0.9503,
"step": 1665
},
{
"epoch": 11.438356164383562,
"grad_norm": 0.423828125,
"learning_rate": 3.242629117644229e-05,
"loss": 0.9564,
"step": 1670
},
{
"epoch": 11.472602739726028,
"grad_norm": 0.41796875,
"learning_rate": 3.184097447195732e-05,
"loss": 0.9714,
"step": 1675
},
{
"epoch": 11.506849315068493,
"grad_norm": 0.404296875,
"learning_rate": 3.125998676740987e-05,
"loss": 0.9516,
"step": 1680
},
{
"epoch": 11.54109589041096,
"grad_norm": 0.416015625,
"learning_rate": 3.068336496320631e-05,
"loss": 0.9634,
"step": 1685
},
{
"epoch": 11.575342465753424,
"grad_norm": 0.419921875,
"learning_rate": 3.0111145682460507e-05,
"loss": 0.9578,
"step": 1690
},
{
"epoch": 11.60958904109589,
"grad_norm": 0.412109375,
"learning_rate": 2.9543365268667867e-05,
"loss": 0.9551,
"step": 1695
},
{
"epoch": 11.643835616438356,
"grad_norm": 0.404296875,
"learning_rate": 2.8980059783396953e-05,
"loss": 0.9585,
"step": 1700
},
{
"epoch": 11.678082191780822,
"grad_norm": 0.390625,
"learning_rate": 2.8421265003999286e-05,
"loss": 0.9573,
"step": 1705
},
{
"epoch": 11.712328767123287,
"grad_norm": 0.4140625,
"learning_rate": 2.7867016421336776e-05,
"loss": 0.9544,
"step": 1710
},
{
"epoch": 11.746575342465754,
"grad_norm": 0.396484375,
"learning_rate": 2.7317349237527724e-05,
"loss": 0.9432,
"step": 1715
},
{
"epoch": 11.780821917808218,
"grad_norm": 0.41015625,
"learning_rate": 2.6772298363710956e-05,
"loss": 0.9567,
"step": 1720
},
{
"epoch": 11.815068493150685,
"grad_norm": 0.431640625,
"learning_rate": 2.6231898417828603e-05,
"loss": 0.9667,
"step": 1725
},
{
"epoch": 11.849315068493151,
"grad_norm": 0.421875,
"learning_rate": 2.569618372242727e-05,
"loss": 0.962,
"step": 1730
},
{
"epoch": 11.883561643835616,
"grad_norm": 0.427734375,
"learning_rate": 2.5165188302478215e-05,
"loss": 0.9514,
"step": 1735
},
{
"epoch": 11.917808219178083,
"grad_norm": 0.435546875,
"learning_rate": 2.4638945883216235e-05,
"loss": 0.9446,
"step": 1740
},
{
"epoch": 11.952054794520548,
"grad_norm": 0.412109375,
"learning_rate": 2.411748988799769e-05,
"loss": 0.9473,
"step": 1745
},
{
"epoch": 11.986301369863014,
"grad_norm": 0.4375,
"learning_rate": 2.3600853436177672e-05,
"loss": 0.97,
"step": 1750
},
{
"epoch": 12.0,
"eval_loss": 2.5957212448120117,
"eval_runtime": 0.5471,
"eval_samples_per_second": 18.279,
"eval_steps_per_second": 1.828,
"step": 1752
},
{
"epoch": 12.020547945205479,
"grad_norm": 0.388671875,
"learning_rate": 2.3089069341006565e-05,
"loss": 0.9617,
"step": 1755
},
{
"epoch": 12.054794520547945,
"grad_norm": 0.447265625,
"learning_rate": 2.2582170107545852e-05,
"loss": 0.9448,
"step": 1760
},
{
"epoch": 12.08904109589041,
"grad_norm": 0.41015625,
"learning_rate": 2.2080187930603668e-05,
"loss": 0.9451,
"step": 1765
},
{
"epoch": 12.123287671232877,
"grad_norm": 0.41796875,
"learning_rate": 2.1583154692689976e-05,
"loss": 0.9472,
"step": 1770
},
{
"epoch": 12.157534246575343,
"grad_norm": 0.412109375,
"learning_rate": 2.109110196199171e-05,
"loss": 0.9536,
"step": 1775
},
{
"epoch": 12.191780821917808,
"grad_norm": 0.3984375,
"learning_rate": 2.0604060990367624e-05,
"loss": 0.9578,
"step": 1780
},
{
"epoch": 12.226027397260275,
"grad_norm": 0.412109375,
"learning_rate": 2.0122062711363532e-05,
"loss": 0.9515,
"step": 1785
},
{
"epoch": 12.26027397260274,
"grad_norm": 0.423828125,
"learning_rate": 1.9645137738247422e-05,
"loss": 0.9438,
"step": 1790
},
{
"epoch": 12.294520547945206,
"grad_norm": 0.40625,
"learning_rate": 1.9173316362065384e-05,
"loss": 0.9516,
"step": 1795
},
{
"epoch": 12.32876712328767,
"grad_norm": 0.41796875,
"learning_rate": 1.8706628549717452e-05,
"loss": 0.9517,
"step": 1800
},
{
"epoch": 12.363013698630137,
"grad_norm": 0.41015625,
"learning_rate": 1.824510394205453e-05,
"loss": 0.9475,
"step": 1805
},
{
"epoch": 12.397260273972602,
"grad_norm": 0.412109375,
"learning_rate": 1.7788771851995655e-05,
"loss": 0.9536,
"step": 1810
},
{
"epoch": 12.431506849315069,
"grad_norm": 0.421875,
"learning_rate": 1.7337661262666294e-05,
"loss": 0.9551,
"step": 1815
},
{
"epoch": 12.465753424657533,
"grad_norm": 0.40625,
"learning_rate": 1.6891800825557535e-05,
"loss": 0.9537,
"step": 1820
},
{
"epoch": 12.5,
"grad_norm": 0.419921875,
"learning_rate": 1.6451218858706374e-05,
"loss": 0.9516,
"step": 1825
},
{
"epoch": 12.534246575342467,
"grad_norm": 0.408203125,
"learning_rate": 1.601594334489702e-05,
"loss": 0.9535,
"step": 1830
},
{
"epoch": 12.568493150684931,
"grad_norm": 0.4375,
"learning_rate": 1.5586001929883865e-05,
"loss": 0.9489,
"step": 1835
},
{
"epoch": 12.602739726027398,
"grad_norm": 0.41015625,
"learning_rate": 1.516142192063521e-05,
"loss": 0.9554,
"step": 1840
},
{
"epoch": 12.636986301369863,
"grad_norm": 0.41015625,
"learning_rate": 1.474223028359939e-05,
"loss": 0.9518,
"step": 1845
},
{
"epoch": 12.67123287671233,
"grad_norm": 0.41015625,
"learning_rate": 1.4328453642991646e-05,
"loss": 0.9565,
"step": 1850
},
{
"epoch": 12.705479452054794,
"grad_norm": 0.396484375,
"learning_rate": 1.392011827910341e-05,
"loss": 0.9538,
"step": 1855
},
{
"epoch": 12.73972602739726,
"grad_norm": 0.416015625,
"learning_rate": 1.3517250126632986e-05,
"loss": 0.9498,
"step": 1860
},
{
"epoch": 12.773972602739725,
"grad_norm": 0.39453125,
"learning_rate": 1.311987477303842e-05,
"loss": 0.9468,
"step": 1865
},
{
"epoch": 12.808219178082192,
"grad_norm": 0.40234375,
"learning_rate": 1.2728017456912344e-05,
"loss": 0.9587,
"step": 1870
},
{
"epoch": 12.842465753424658,
"grad_norm": 0.419921875,
"learning_rate": 1.2341703066379074e-05,
"loss": 0.9463,
"step": 1875
},
{
"epoch": 12.876712328767123,
"grad_norm": 0.3984375,
"learning_rate": 1.1960956137513701e-05,
"loss": 0.9506,
"step": 1880
},
{
"epoch": 12.91095890410959,
"grad_norm": 0.408203125,
"learning_rate": 1.158580085278398e-05,
"loss": 0.9492,
"step": 1885
},
{
"epoch": 12.945205479452055,
"grad_norm": 0.416015625,
"learning_rate": 1.1216261039514087e-05,
"loss": 0.9569,
"step": 1890
},
{
"epoch": 12.979452054794521,
"grad_norm": 0.408203125,
"learning_rate": 1.0852360168371656e-05,
"loss": 0.9514,
"step": 1895
},
{
"epoch": 13.0,
"eval_loss": 2.598764181137085,
"eval_runtime": 0.5483,
"eval_samples_per_second": 18.237,
"eval_steps_per_second": 1.824,
"step": 1898
},
{
"epoch": 13.013698630136986,
"grad_norm": 0.412109375,
"learning_rate": 1.049412135187675e-05,
"loss": 0.9556,
"step": 1900
},
{
"epoch": 13.047945205479452,
"grad_norm": 0.40234375,
"learning_rate": 1.0141567342934132e-05,
"loss": 0.9543,
"step": 1905
},
{
"epoch": 13.082191780821917,
"grad_norm": 0.423828125,
"learning_rate": 9.794720533388024e-06,
"loss": 0.947,
"step": 1910
},
{
"epoch": 13.116438356164384,
"grad_norm": 0.408203125,
"learning_rate": 9.453602952599982e-06,
"loss": 0.9567,
"step": 1915
},
{
"epoch": 13.150684931506849,
"grad_norm": 0.408203125,
"learning_rate": 9.118236266049707e-06,
"loss": 0.956,
"step": 1920
},
{
"epoch": 13.184931506849315,
"grad_norm": 0.41015625,
"learning_rate": 8.788641773959105e-06,
"loss": 0.9508,
"step": 1925
},
{
"epoch": 13.219178082191782,
"grad_norm": 0.419921875,
"learning_rate": 8.464840409939267e-06,
"loss": 0.9556,
"step": 1930
},
{
"epoch": 13.253424657534246,
"grad_norm": 0.41015625,
"learning_rate": 8.146852739661105e-06,
"loss": 0.9577,
"step": 1935
},
{
"epoch": 13.287671232876713,
"grad_norm": 0.400390625,
"learning_rate": 7.834698959548914e-06,
"loss": 0.9505,
"step": 1940
},
{
"epoch": 13.321917808219178,
"grad_norm": 0.404296875,
"learning_rate": 7.528398895497924e-06,
"loss": 0.9556,
"step": 1945
},
{
"epoch": 13.356164383561644,
"grad_norm": 0.400390625,
"learning_rate": 7.2279720016148244e-06,
"loss": 0.9445,
"step": 1950
},
{
"epoch": 13.39041095890411,
"grad_norm": 0.40625,
"learning_rate": 6.933437358982409e-06,
"loss": 0.9491,
"step": 1955
},
{
"epoch": 13.424657534246576,
"grad_norm": 0.3984375,
"learning_rate": 6.6448136744474474e-06,
"loss": 0.9528,
"step": 1960
},
{
"epoch": 13.45890410958904,
"grad_norm": 0.40234375,
"learning_rate": 6.36211927943271e-06,
"loss": 0.9531,
"step": 1965
},
{
"epoch": 13.493150684931507,
"grad_norm": 0.43359375,
"learning_rate": 6.085372128772637e-06,
"loss": 0.949,
"step": 1970
},
{
"epoch": 13.527397260273972,
"grad_norm": 0.408203125,
"learning_rate": 5.814589799572956e-06,
"loss": 0.9538,
"step": 1975
},
{
"epoch": 13.561643835616438,
"grad_norm": 0.40625,
"learning_rate": 5.549789490094304e-06,
"loss": 0.9481,
"step": 1980
},
{
"epoch": 13.595890410958905,
"grad_norm": 0.412109375,
"learning_rate": 5.290988018659937e-06,
"loss": 0.9523,
"step": 1985
},
{
"epoch": 13.63013698630137,
"grad_norm": 0.390625,
"learning_rate": 5.038201822587474e-06,
"loss": 0.9483,
"step": 1990
},
{
"epoch": 13.664383561643836,
"grad_norm": 0.416015625,
"learning_rate": 4.79144695714504e-06,
"loss": 0.9478,
"step": 1995
},
{
"epoch": 13.698630136986301,
"grad_norm": 0.404296875,
"learning_rate": 4.550739094531386e-06,
"loss": 0.9459,
"step": 2000
},
{
"epoch": 13.732876712328768,
"grad_norm": 0.38671875,
"learning_rate": 4.316093522880648e-06,
"loss": 0.9474,
"step": 2005
},
{
"epoch": 13.767123287671232,
"grad_norm": 0.392578125,
"learning_rate": 4.087525145291204e-06,
"loss": 0.9457,
"step": 2010
},
{
"epoch": 13.801369863013699,
"grad_norm": 0.408203125,
"learning_rate": 3.865048478879241e-06,
"loss": 0.9483,
"step": 2015
},
{
"epoch": 13.835616438356164,
"grad_norm": 0.404296875,
"learning_rate": 3.6486776538566803e-06,
"loss": 0.9491,
"step": 2020
},
{
"epoch": 13.86986301369863,
"grad_norm": 0.41796875,
"learning_rate": 3.4384264126337328e-06,
"loss": 0.9497,
"step": 2025
},
{
"epoch": 13.904109589041095,
"grad_norm": 0.396484375,
"learning_rate": 3.2343081089460603e-06,
"loss": 0.9514,
"step": 2030
},
{
"epoch": 13.938356164383562,
"grad_norm": 0.40234375,
"learning_rate": 3.0363357070066544e-06,
"loss": 0.9484,
"step": 2035
},
{
"epoch": 13.972602739726028,
"grad_norm": 0.431640625,
"learning_rate": 2.8445217806824077e-06,
"loss": 0.9469,
"step": 2040
},
{
"epoch": 14.0,
"eval_loss": 2.599691867828369,
"eval_runtime": 0.5385,
"eval_samples_per_second": 18.57,
"eval_steps_per_second": 1.857,
"step": 2044
},
{
"epoch": 14.006849315068493,
"grad_norm": 0.4296875,
"learning_rate": 2.658878512695562e-06,
"loss": 0.9462,
"step": 2045
},
{
"epoch": 14.04109589041096,
"grad_norm": 0.41015625,
"learning_rate": 2.4794176938498837e-06,
"loss": 0.9509,
"step": 2050
},
{
"epoch": 14.075342465753424,
"grad_norm": 0.39453125,
"learning_rate": 2.30615072228183e-06,
"loss": 0.9438,
"step": 2055
},
{
"epoch": 14.10958904109589,
"grad_norm": 0.400390625,
"learning_rate": 2.139088602736616e-06,
"loss": 0.9567,
"step": 2060
},
{
"epoch": 14.143835616438356,
"grad_norm": 0.404296875,
"learning_rate": 1.9782419458692193e-06,
"loss": 0.9454,
"step": 2065
},
{
"epoch": 14.178082191780822,
"grad_norm": 0.404296875,
"learning_rate": 1.8236209675705274e-06,
"loss": 0.9486,
"step": 2070
},
{
"epoch": 14.212328767123287,
"grad_norm": 0.408203125,
"learning_rate": 1.6752354883184717e-06,
"loss": 0.9521,
"step": 2075
},
{
"epoch": 14.246575342465754,
"grad_norm": 0.40625,
"learning_rate": 1.5330949325542797e-06,
"loss": 0.9556,
"step": 2080
},
{
"epoch": 14.280821917808218,
"grad_norm": 0.408203125,
"learning_rate": 1.397208328083921e-06,
"loss": 0.9574,
"step": 2085
},
{
"epoch": 14.315068493150685,
"grad_norm": 0.427734375,
"learning_rate": 1.2675843055046765e-06,
"loss": 0.9557,
"step": 2090
},
{
"epoch": 14.349315068493151,
"grad_norm": 0.41015625,
"learning_rate": 1.144231097657078e-06,
"loss": 0.952,
"step": 2095
},
{
"epoch": 14.383561643835616,
"grad_norm": 0.404296875,
"learning_rate": 1.0271565391018922e-06,
"loss": 0.9475,
"step": 2100
},
{
"epoch": 14.417808219178083,
"grad_norm": 0.41796875,
"learning_rate": 9.163680656226303e-07,
"loss": 0.9548,
"step": 2105
},
{
"epoch": 14.452054794520548,
"grad_norm": 0.404296875,
"learning_rate": 8.118727137532034e-07,
"loss": 0.9473,
"step": 2110
},
{
"epoch": 14.486301369863014,
"grad_norm": 0.396484375,
"learning_rate": 7.136771203310245e-07,
"loss": 0.9593,
"step": 2115
},
{
"epoch": 14.520547945205479,
"grad_norm": 0.404296875,
"learning_rate": 6.21787522075512e-07,
"loss": 0.9442,
"step": 2120
},
{
"epoch": 14.554794520547945,
"grad_norm": 0.396484375,
"learning_rate": 5.362097551919631e-07,
"loss": 0.9509,
"step": 2125
},
{
"epoch": 14.58904109589041,
"grad_norm": 0.408203125,
"learning_rate": 4.569492550008603e-07,
"loss": 0.943,
"step": 2130
},
{
"epoch": 14.623287671232877,
"grad_norm": 0.408203125,
"learning_rate": 3.84011055592659e-07,
"loss": 0.9539,
"step": 2135
},
{
"epoch": 14.657534246575342,
"grad_norm": 0.4140625,
"learning_rate": 3.1739978950806603e-07,
"loss": 0.9477,
"step": 2140
},
{
"epoch": 14.691780821917808,
"grad_norm": 0.404296875,
"learning_rate": 2.5711968744382974e-07,
"loss": 0.9532,
"step": 2145
},
{
"epoch": 14.726027397260275,
"grad_norm": 0.404296875,
"learning_rate": 2.0317457798398888e-07,
"loss": 0.9453,
"step": 2150
},
{
"epoch": 14.76027397260274,
"grad_norm": 0.392578125,
"learning_rate": 1.5556788735676675e-07,
"loss": 0.9501,
"step": 2155
},
{
"epoch": 14.794520547945206,
"grad_norm": 0.423828125,
"learning_rate": 1.143026392168789e-07,
"loss": 0.9372,
"step": 2160
},
{
"epoch": 14.82876712328767,
"grad_norm": 0.39453125,
"learning_rate": 7.938145445357536e-08,
"loss": 0.9602,
"step": 2165
},
{
"epoch": 14.863013698630137,
"grad_norm": 0.41015625,
"learning_rate": 5.0806551024129565e-08,
"loss": 0.9451,
"step": 2170
},
{
"epoch": 14.897260273972602,
"grad_norm": 0.412109375,
"learning_rate": 2.8579743813006432e-08,
"loss": 0.9433,
"step": 2175
},
{
"epoch": 14.931506849315069,
"grad_norm": 0.408203125,
"learning_rate": 1.270244451652136e-08,
"loss": 0.9548,
"step": 2180
},
{
"epoch": 14.965753424657533,
"grad_norm": 0.408203125,
"learning_rate": 3.175661553256326e-09,
"loss": 0.9541,
"step": 2185
},
{
"epoch": 15.0,
"grad_norm": 0.408203125,
"learning_rate": 0.0,
"loss": 0.9469,
"step": 2190
},
{
"epoch": 15.0,
"eval_loss": 2.5990421772003174,
"eval_runtime": 0.5553,
"eval_samples_per_second": 18.009,
"eval_steps_per_second": 1.801,
"step": 2190
},
{
"epoch": 15.0,
"step": 2190,
"total_flos": 1.2863476116823736e+18,
"train_loss": 1.080002195214572,
"train_runtime": 11705.6736,
"train_samples_per_second": 8.973,
"train_steps_per_second": 0.187
}
],
"logging_steps": 5,
"max_steps": 2190,
"num_input_tokens_seen": 0,
"num_train_epochs": 15,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 1.2863476116823736e+18,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}