3N3G's picture
End of training
b1adf1a verified
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 12.0,
"eval_steps": 500,
"global_step": 108,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.11490125673249552,
"grad_norm": 1.8050661087036133,
"learning_rate": 0.0,
"loss": 0.2697,
"step": 1
},
{
"epoch": 0.22980251346499103,
"grad_norm": 1.9665236473083496,
"learning_rate": 9.09090909090909e-08,
"loss": 0.2879,
"step": 2
},
{
"epoch": 0.34470377019748655,
"grad_norm": 1.8855077028274536,
"learning_rate": 1.818181818181818e-07,
"loss": 0.2696,
"step": 3
},
{
"epoch": 0.45960502692998206,
"grad_norm": 1.998793125152588,
"learning_rate": 2.727272727272727e-07,
"loss": 0.3002,
"step": 4
},
{
"epoch": 0.5745062836624776,
"grad_norm": 1.83466637134552,
"learning_rate": 3.636363636363636e-07,
"loss": 0.2759,
"step": 5
},
{
"epoch": 0.6894075403949731,
"grad_norm": 1.848267912864685,
"learning_rate": 4.545454545454545e-07,
"loss": 0.2759,
"step": 6
},
{
"epoch": 0.8043087971274686,
"grad_norm": 1.7061188220977783,
"learning_rate": 5.454545454545454e-07,
"loss": 0.2666,
"step": 7
},
{
"epoch": 0.9192100538599641,
"grad_norm": 1.8305636644363403,
"learning_rate": 6.363636363636363e-07,
"loss": 0.2805,
"step": 8
},
{
"epoch": 1.0,
"grad_norm": 1.634291172027588,
"learning_rate": 7.272727272727272e-07,
"loss": 0.2622,
"step": 9
},
{
"epoch": 1.1149012567324954,
"grad_norm": 2.2485928535461426,
"learning_rate": 8.181818181818182e-07,
"loss": 0.2544,
"step": 10
},
{
"epoch": 1.229802513464991,
"grad_norm": 1.6178233623504639,
"learning_rate": 9.09090909090909e-07,
"loss": 0.2777,
"step": 11
},
{
"epoch": 1.3447037701974867,
"grad_norm": 1.6333025693893433,
"learning_rate": 1e-06,
"loss": 0.2825,
"step": 12
},
{
"epoch": 1.459605026929982,
"grad_norm": 1.5703474283218384,
"learning_rate": 9.997640060704816e-07,
"loss": 0.2715,
"step": 13
},
{
"epoch": 1.5745062836624775,
"grad_norm": 1.486207127571106,
"learning_rate": 9.990562718069702e-07,
"loss": 0.283,
"step": 14
},
{
"epoch": 1.689407540394973,
"grad_norm": 1.2863779067993164,
"learning_rate": 9.978775395249762e-07,
"loss": 0.2554,
"step": 15
},
{
"epoch": 1.8043087971274687,
"grad_norm": 1.2911548614501953,
"learning_rate": 9.962290455518912e-07,
"loss": 0.2629,
"step": 16
},
{
"epoch": 1.9192100538599641,
"grad_norm": 1.2554247379302979,
"learning_rate": 9.941125189302508e-07,
"loss": 0.2486,
"step": 17
},
{
"epoch": 2.0,
"grad_norm": 1.239845871925354,
"learning_rate": 9.915301796042075e-07,
"loss": 0.2368,
"step": 18
},
{
"epoch": 2.1149012567324954,
"grad_norm": 1.3584400415420532,
"learning_rate": 9.884847360911167e-07,
"loss": 0.2628,
"step": 19
},
{
"epoch": 2.229802513464991,
"grad_norm": 1.2527025938034058,
"learning_rate": 9.84979382640675e-07,
"loss": 0.2419,
"step": 20
},
{
"epoch": 2.3447037701974867,
"grad_norm": 1.2097444534301758,
"learning_rate": 9.81017795884594e-07,
"loss": 0.2394,
"step": 21
},
{
"epoch": 2.459605026929982,
"grad_norm": 1.0766065120697021,
"learning_rate": 9.766041309803217e-07,
"loss": 0.2463,
"step": 22
},
{
"epoch": 2.5745062836624775,
"grad_norm": 0.922896146774292,
"learning_rate": 9.717430172528546e-07,
"loss": 0.2435,
"step": 23
},
{
"epoch": 2.6894075403949733,
"grad_norm": 0.8510429263114929,
"learning_rate": 9.66439553339217e-07,
"loss": 0.2445,
"step": 24
},
{
"epoch": 2.8043087971274687,
"grad_norm": 0.8421075344085693,
"learning_rate": 9.60699301840693e-07,
"loss": 0.222,
"step": 25
},
{
"epoch": 2.919210053859964,
"grad_norm": 0.8339232802391052,
"learning_rate": 9.54528283488428e-07,
"loss": 0.2197,
"step": 26
},
{
"epoch": 3.0,
"grad_norm": 1.2300190925598145,
"learning_rate": 9.479329708285106e-07,
"loss": 0.2308,
"step": 27
},
{
"epoch": 3.1149012567324954,
"grad_norm": 0.8397951126098633,
"learning_rate": 9.409202814331679e-07,
"loss": 0.2355,
"step": 28
},
{
"epoch": 3.229802513464991,
"grad_norm": 0.774864137172699,
"learning_rate": 9.334975706451861e-07,
"loss": 0.2316,
"step": 29
},
{
"epoch": 3.3447037701974867,
"grad_norm": 0.8159404993057251,
"learning_rate": 9.256726238631719e-07,
"loss": 0.2309,
"step": 30
},
{
"epoch": 3.459605026929982,
"grad_norm": 0.6702387928962708,
"learning_rate": 9.174536483757448e-07,
"loss": 0.2132,
"step": 31
},
{
"epoch": 3.5745062836624775,
"grad_norm": 0.6168584823608398,
"learning_rate": 9.088492647532243e-07,
"loss": 0.2064,
"step": 32
},
{
"epoch": 3.6894075403949733,
"grad_norm": 0.6055068373680115,
"learning_rate": 8.998684978058422e-07,
"loss": 0.2158,
"step": 33
},
{
"epoch": 3.8043087971274687,
"grad_norm": 0.6474173665046692,
"learning_rate": 8.905207671179627e-07,
"loss": 0.225,
"step": 34
},
{
"epoch": 3.919210053859964,
"grad_norm": 0.5831088423728943,
"learning_rate": 8.808158771682401e-07,
"loss": 0.2271,
"step": 35
},
{
"epoch": 4.0,
"grad_norm": 0.7952995896339417,
"learning_rate": 8.707640070460731e-07,
"loss": 0.1925,
"step": 36
},
{
"epoch": 4.114901256732495,
"grad_norm": 0.5484858751296997,
"learning_rate": 8.60375699775147e-07,
"loss": 0.2108,
"step": 37
},
{
"epoch": 4.229802513464991,
"grad_norm": 0.5406907200813293,
"learning_rate": 8.496618512552564e-07,
"loss": 0.2082,
"step": 38
},
{
"epoch": 4.344703770197486,
"grad_norm": 0.6845734119415283,
"learning_rate": 8.386336988340129e-07,
"loss": 0.2266,
"step": 39
},
{
"epoch": 4.459605026929982,
"grad_norm": 0.8074614405632019,
"learning_rate": 8.273028095204173e-07,
"loss": 0.2342,
"step": 40
},
{
"epoch": 4.574506283662478,
"grad_norm": 0.7541784048080444,
"learning_rate": 8.156810678526652e-07,
"loss": 0.2178,
"step": 41
},
{
"epoch": 4.689407540394973,
"grad_norm": 0.6619231700897217,
"learning_rate": 8.037806634329078e-07,
"loss": 0.1966,
"step": 42
},
{
"epoch": 4.804308797127469,
"grad_norm": 0.6785462498664856,
"learning_rate": 7.916140781420428e-07,
"loss": 0.2001,
"step": 43
},
{
"epoch": 4.919210053859964,
"grad_norm": 0.6341767907142639,
"learning_rate": 7.791940730479434e-07,
"loss": 0.1885,
"step": 44
},
{
"epoch": 5.0,
"grad_norm": 0.6831431984901428,
"learning_rate": 7.665336750208623e-07,
"loss": 0.2095,
"step": 45
},
{
"epoch": 5.114901256732495,
"grad_norm": 0.5810871720314026,
"learning_rate": 7.536461630700425e-07,
"loss": 0.2121,
"step": 46
},
{
"epoch": 5.229802513464991,
"grad_norm": 0.49919337034225464,
"learning_rate": 7.405450544158706e-07,
"loss": 0.1943,
"step": 47
},
{
"epoch": 5.344703770197486,
"grad_norm": 0.5304792523384094,
"learning_rate": 7.272440903121791e-07,
"loss": 0.2043,
"step": 48
},
{
"epoch": 5.459605026929982,
"grad_norm": 0.5594088435173035,
"learning_rate": 7.137572216335694e-07,
"loss": 0.2177,
"step": 49
},
{
"epoch": 5.574506283662478,
"grad_norm": 0.7475131154060364,
"learning_rate": 7.000985942428693e-07,
"loss": 0.1891,
"step": 50
},
{
"epoch": 5.689407540394973,
"grad_norm": 0.554442822933197,
"learning_rate": 6.862825341540778e-07,
"loss": 0.1867,
"step": 51
},
{
"epoch": 5.804308797127469,
"grad_norm": 0.5080501437187195,
"learning_rate": 6.723235325063543e-07,
"loss": 0.2069,
"step": 52
},
{
"epoch": 5.919210053859964,
"grad_norm": 0.6186196804046631,
"learning_rate": 6.582362303648142e-07,
"loss": 0.2093,
"step": 53
},
{
"epoch": 6.0,
"grad_norm": 0.6012147665023804,
"learning_rate": 6.440354033640738e-07,
"loss": 0.1712,
"step": 54
},
{
"epoch": 6.114901256732495,
"grad_norm": 0.5101262927055359,
"learning_rate": 6.297359462106502e-07,
"loss": 0.1974,
"step": 55
},
{
"epoch": 6.229802513464991,
"grad_norm": 0.4870443642139435,
"learning_rate": 6.153528570604707e-07,
"loss": 0.2023,
"step": 56
},
{
"epoch": 6.344703770197486,
"grad_norm": 0.46721193194389343,
"learning_rate": 6.00901221787878e-07,
"loss": 0.1897,
"step": 57
},
{
"epoch": 6.459605026929982,
"grad_norm": 0.4450569152832031,
"learning_rate": 5.86396198162632e-07,
"loss": 0.1932,
"step": 58
},
{
"epoch": 6.574506283662478,
"grad_norm": 0.4395124912261963,
"learning_rate": 5.718529999515017e-07,
"loss": 0.1851,
"step": 59
},
{
"epoch": 6.689407540394973,
"grad_norm": 0.4729479253292084,
"learning_rate": 5.572868809611257e-07,
"loss": 0.2042,
"step": 60
},
{
"epoch": 6.804308797127469,
"grad_norm": 0.4454880952835083,
"learning_rate": 5.427131190388743e-07,
"loss": 0.1925,
"step": 61
},
{
"epoch": 6.919210053859964,
"grad_norm": 0.40814733505249023,
"learning_rate": 5.281470000484985e-07,
"loss": 0.1776,
"step": 62
},
{
"epoch": 7.0,
"grad_norm": 0.48163506388664246,
"learning_rate": 5.136038018373682e-07,
"loss": 0.2039,
"step": 63
},
{
"epoch": 7.114901256732495,
"grad_norm": 0.4723432660102844,
"learning_rate": 4.990987782121221e-07,
"loss": 0.2026,
"step": 64
},
{
"epoch": 7.229802513464991,
"grad_norm": 0.4103083610534668,
"learning_rate": 4.846471429395295e-07,
"loss": 0.184,
"step": 65
},
{
"epoch": 7.344703770197486,
"grad_norm": 0.4164697527885437,
"learning_rate": 4.7026405378934975e-07,
"loss": 0.1951,
"step": 66
},
{
"epoch": 7.459605026929982,
"grad_norm": 0.43192094564437866,
"learning_rate": 4.5596459663592625e-07,
"loss": 0.1888,
"step": 67
},
{
"epoch": 7.574506283662478,
"grad_norm": 0.40324753522872925,
"learning_rate": 4.41763769635186e-07,
"loss": 0.1969,
"step": 68
},
{
"epoch": 7.689407540394973,
"grad_norm": 0.4374394416809082,
"learning_rate": 4.2767646749364574e-07,
"loss": 0.1801,
"step": 69
},
{
"epoch": 7.804308797127469,
"grad_norm": 0.40512681007385254,
"learning_rate": 4.1371746584592227e-07,
"loss": 0.1849,
"step": 70
},
{
"epoch": 7.919210053859964,
"grad_norm": 0.40997177362442017,
"learning_rate": 3.999014057571308e-07,
"loss": 0.1922,
"step": 71
},
{
"epoch": 8.0,
"grad_norm": 0.48699864745140076,
"learning_rate": 3.862427783664306e-07,
"loss": 0.1774,
"step": 72
},
{
"epoch": 8.114901256732496,
"grad_norm": 0.4171445965766907,
"learning_rate": 3.7275590968782087e-07,
"loss": 0.1887,
"step": 73
},
{
"epoch": 8.22980251346499,
"grad_norm": 0.3887261152267456,
"learning_rate": 3.594549455841296e-07,
"loss": 0.183,
"step": 74
},
{
"epoch": 8.344703770197487,
"grad_norm": 0.41559508442878723,
"learning_rate": 3.4635383692995755e-07,
"loss": 0.1858,
"step": 75
},
{
"epoch": 8.459605026929982,
"grad_norm": 0.41108807921409607,
"learning_rate": 3.3346632497913773e-07,
"loss": 0.1818,
"step": 76
},
{
"epoch": 8.574506283662478,
"grad_norm": 0.4802042245864868,
"learning_rate": 3.208059269520568e-07,
"loss": 0.1952,
"step": 77
},
{
"epoch": 8.689407540394972,
"grad_norm": 0.3916759192943573,
"learning_rate": 3.083859218579573e-07,
"loss": 0.181,
"step": 78
},
{
"epoch": 8.804308797127469,
"grad_norm": 0.42845413088798523,
"learning_rate": 2.9621933656709207e-07,
"loss": 0.1948,
"step": 79
},
{
"epoch": 8.919210053859963,
"grad_norm": 0.4223984181880951,
"learning_rate": 2.843189321473349e-07,
"loss": 0.1951,
"step": 80
},
{
"epoch": 9.0,
"grad_norm": 0.44743412733078003,
"learning_rate": 2.7269719047958267e-07,
"loss": 0.1727,
"step": 81
},
{
"epoch": 9.114901256732496,
"grad_norm": 0.41905954480171204,
"learning_rate": 2.613663011659871e-07,
"loss": 0.1757,
"step": 82
},
{
"epoch": 9.22980251346499,
"grad_norm": 0.399945467710495,
"learning_rate": 2.5033814874474356e-07,
"loss": 0.1937,
"step": 83
},
{
"epoch": 9.344703770197487,
"grad_norm": 0.3881186842918396,
"learning_rate": 2.3962430022485305e-07,
"loss": 0.1761,
"step": 84
},
{
"epoch": 9.459605026929982,
"grad_norm": 0.3914501368999481,
"learning_rate": 2.2923599295392694e-07,
"loss": 0.19,
"step": 85
},
{
"epoch": 9.574506283662478,
"grad_norm": 0.41606462001800537,
"learning_rate": 2.1918412283175994e-07,
"loss": 0.189,
"step": 86
},
{
"epoch": 9.689407540394972,
"grad_norm": 0.3928024470806122,
"learning_rate": 2.0947923288203713e-07,
"loss": 0.1857,
"step": 87
},
{
"epoch": 9.804308797127469,
"grad_norm": 0.3906983435153961,
"learning_rate": 2.0013150219415793e-07,
"loss": 0.1822,
"step": 88
},
{
"epoch": 9.919210053859963,
"grad_norm": 0.3903934061527252,
"learning_rate": 1.9115073524677572e-07,
"loss": 0.1856,
"step": 89
},
{
"epoch": 10.0,
"grad_norm": 0.5587829947471619,
"learning_rate": 1.8254635162425503e-07,
"loss": 0.1863,
"step": 90
},
{
"epoch": 10.114901256732496,
"grad_norm": 0.3936590850353241,
"learning_rate": 1.7432737613682807e-07,
"loss": 0.1954,
"step": 91
},
{
"epoch": 10.22980251346499,
"grad_norm": 0.3949429988861084,
"learning_rate": 1.6650242935481388e-07,
"loss": 0.1861,
"step": 92
},
{
"epoch": 10.344703770197487,
"grad_norm": 0.4262572228908539,
"learning_rate": 1.59079718566832e-07,
"loss": 0.198,
"step": 93
},
{
"epoch": 10.459605026929982,
"grad_norm": 0.37948721647262573,
"learning_rate": 1.5206702917148945e-07,
"loss": 0.1711,
"step": 94
},
{
"epoch": 10.574506283662478,
"grad_norm": 0.3690933585166931,
"learning_rate": 1.4547171651157214e-07,
"loss": 0.1642,
"step": 95
},
{
"epoch": 10.689407540394972,
"grad_norm": 0.3985244333744049,
"learning_rate": 1.3930069815930697e-07,
"loss": 0.1866,
"step": 96
},
{
"epoch": 10.804308797127469,
"grad_norm": 0.3972547650337219,
"learning_rate": 1.3356044666078315e-07,
"loss": 0.1816,
"step": 97
},
{
"epoch": 10.919210053859963,
"grad_norm": 0.403357595205307,
"learning_rate": 1.2825698274714542e-07,
"loss": 0.1804,
"step": 98
},
{
"epoch": 11.0,
"grad_norm": 0.49961942434310913,
"learning_rate": 1.233958690196783e-07,
"loss": 0.192,
"step": 99
},
{
"epoch": 11.114901256732496,
"grad_norm": 0.38747331500053406,
"learning_rate": 1.1898220411540583e-07,
"loss": 0.1722,
"step": 100
},
{
"epoch": 11.22980251346499,
"grad_norm": 0.41416335105895996,
"learning_rate": 1.1502061735932499e-07,
"loss": 0.1891,
"step": 101
},
{
"epoch": 11.344703770197487,
"grad_norm": 0.41525566577911377,
"learning_rate": 1.115152639088833e-07,
"loss": 0.1732,
"step": 102
},
{
"epoch": 11.459605026929982,
"grad_norm": 0.39968863129615784,
"learning_rate": 1.0846982039579242e-07,
"loss": 0.191,
"step": 103
},
{
"epoch": 11.574506283662478,
"grad_norm": 0.4091079831123352,
"learning_rate": 1.0588748106974918e-07,
"loss": 0.1959,
"step": 104
},
{
"epoch": 11.689407540394972,
"grad_norm": 0.38746750354766846,
"learning_rate": 1.0377095444810871e-07,
"loss": 0.1811,
"step": 105
},
{
"epoch": 11.804308797127469,
"grad_norm": 0.3729393184185028,
"learning_rate": 1.0212246047502372e-07,
"loss": 0.1757,
"step": 106
},
{
"epoch": 11.919210053859963,
"grad_norm": 0.42114880681037903,
"learning_rate": 1.0094372819302977e-07,
"loss": 0.1775,
"step": 107
},
{
"epoch": 12.0,
"grad_norm": 0.5820653438568115,
"learning_rate": 1.0023599392951829e-07,
"loss": 0.198,
"step": 108
},
{
"epoch": 12.0,
"step": 108,
"total_flos": 8.159286503861453e+17,
"train_loss": 0.21086449852144276,
"train_runtime": 7716.2568,
"train_samples_per_second": 0.866,
"train_steps_per_second": 0.014
}
],
"logging_steps": 1,
"max_steps": 108,
"num_input_tokens_seen": 0,
"num_train_epochs": 12,
"save_steps": 64000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 8.159286503861453e+17,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}