Romain-XV's picture
Training in progress, step 128, checkpoint
2e81cf3 verified
{
"best_metric": 0.24994200468063354,
"best_model_checkpoint": "miner_id_24/checkpoint-100",
"epoch": 1.006859382655561,
"eval_steps": 50,
"global_step": 128,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.007839294463498285,
"grad_norm": 151.8634796142578,
"learning_rate": 2e-05,
"loss": 21.9839,
"step": 1
},
{
"epoch": 0.007839294463498285,
"eval_loss": 1.2963035106658936,
"eval_runtime": 59.0536,
"eval_samples_per_second": 7.282,
"eval_steps_per_second": 1.829,
"step": 1
},
{
"epoch": 0.01567858892699657,
"grad_norm": 134.68370056152344,
"learning_rate": 4e-05,
"loss": 20.4607,
"step": 2
},
{
"epoch": 0.023517883390494855,
"grad_norm": 106.47999572753906,
"learning_rate": 6e-05,
"loss": 18.8466,
"step": 3
},
{
"epoch": 0.03135717785399314,
"grad_norm": 68.35124969482422,
"learning_rate": 8e-05,
"loss": 14.5929,
"step": 4
},
{
"epoch": 0.039196472317491425,
"grad_norm": 50.851531982421875,
"learning_rate": 0.0001,
"loss": 10.9012,
"step": 5
},
{
"epoch": 0.04703576678098971,
"grad_norm": 37.46467971801758,
"learning_rate": 0.00012,
"loss": 7.959,
"step": 6
},
{
"epoch": 0.054875061244487995,
"grad_norm": 19.023597717285156,
"learning_rate": 0.00014,
"loss": 6.095,
"step": 7
},
{
"epoch": 0.06271435570798628,
"grad_norm": 11.678672790527344,
"learning_rate": 0.00016,
"loss": 5.9509,
"step": 8
},
{
"epoch": 0.07055365017148457,
"grad_norm": 9.556014060974121,
"learning_rate": 0.00018,
"loss": 4.9657,
"step": 9
},
{
"epoch": 0.07839294463498285,
"grad_norm": 9.676369667053223,
"learning_rate": 0.0002,
"loss": 5.2787,
"step": 10
},
{
"epoch": 0.08623223909848114,
"grad_norm": 6.937280178070068,
"learning_rate": 0.00019996456111234527,
"loss": 4.9893,
"step": 11
},
{
"epoch": 0.09407153356197942,
"grad_norm": 6.7998046875,
"learning_rate": 0.0001998582695676762,
"loss": 5.3298,
"step": 12
},
{
"epoch": 0.10191082802547771,
"grad_norm": 6.393218040466309,
"learning_rate": 0.000199681200703075,
"loss": 4.416,
"step": 13
},
{
"epoch": 0.10975012248897599,
"grad_norm": 6.558236122131348,
"learning_rate": 0.00019943348002101371,
"loss": 4.915,
"step": 14
},
{
"epoch": 0.11758941695247428,
"grad_norm": 5.369983196258545,
"learning_rate": 0.00019911528310040074,
"loss": 4.0733,
"step": 15
},
{
"epoch": 0.12542871141597256,
"grad_norm": 6.554808616638184,
"learning_rate": 0.00019872683547213446,
"loss": 5.1536,
"step": 16
},
{
"epoch": 0.13326800587947085,
"grad_norm": 10.400140762329102,
"learning_rate": 0.00019826841245925212,
"loss": 4.4356,
"step": 17
},
{
"epoch": 0.14110730034296914,
"grad_norm": 6.700842380523682,
"learning_rate": 0.00019774033898178667,
"loss": 4.9583,
"step": 18
},
{
"epoch": 0.1489465948064674,
"grad_norm": 7.305096626281738,
"learning_rate": 0.00019714298932647098,
"loss": 4.8479,
"step": 19
},
{
"epoch": 0.1567858892699657,
"grad_norm": 5.645997524261475,
"learning_rate": 0.0001964767868814516,
"loss": 5.104,
"step": 20
},
{
"epoch": 0.164625183733464,
"grad_norm": 5.677758693695068,
"learning_rate": 0.00019574220383620055,
"loss": 4.1287,
"step": 21
},
{
"epoch": 0.17246447819696228,
"grad_norm": 5.749682903289795,
"learning_rate": 0.00019493976084683813,
"loss": 4.3489,
"step": 22
},
{
"epoch": 0.18030377266046055,
"grad_norm": 6.3827948570251465,
"learning_rate": 0.00019407002666710336,
"loss": 4.252,
"step": 23
},
{
"epoch": 0.18814306712395884,
"grad_norm": 5.22667932510376,
"learning_rate": 0.00019313361774523385,
"loss": 4.2697,
"step": 24
},
{
"epoch": 0.19598236158745713,
"grad_norm": 5.128467559814453,
"learning_rate": 0.00019213119778704128,
"loss": 4.1894,
"step": 25
},
{
"epoch": 0.20382165605095542,
"grad_norm": 5.749606132507324,
"learning_rate": 0.00019106347728549135,
"loss": 4.4464,
"step": 26
},
{
"epoch": 0.2116609505144537,
"grad_norm": 5.480006694793701,
"learning_rate": 0.00018993121301712193,
"loss": 4.5645,
"step": 27
},
{
"epoch": 0.21950024497795198,
"grad_norm": 5.424717903137207,
"learning_rate": 0.00018873520750565718,
"loss": 4.6284,
"step": 28
},
{
"epoch": 0.22733953944145027,
"grad_norm": 44.62896728515625,
"learning_rate": 0.00018747630845319612,
"loss": 4.5385,
"step": 29
},
{
"epoch": 0.23517883390494856,
"grad_norm": 6.1519622802734375,
"learning_rate": 0.0001861554081393806,
"loss": 4.7595,
"step": 30
},
{
"epoch": 0.24301812836844683,
"grad_norm": 6.095805644989014,
"learning_rate": 0.0001847734427889671,
"loss": 4.3621,
"step": 31
},
{
"epoch": 0.2508574228319451,
"grad_norm": 5.428371906280518,
"learning_rate": 0.0001833313919082515,
"loss": 4.8596,
"step": 32
},
{
"epoch": 0.2586967172954434,
"grad_norm": 4.395357131958008,
"learning_rate": 0.0001818302775908169,
"loss": 4.1379,
"step": 33
},
{
"epoch": 0.2665360117589417,
"grad_norm": 10.262712478637695,
"learning_rate": 0.00018027116379309638,
"loss": 4.634,
"step": 34
},
{
"epoch": 0.27437530622244,
"grad_norm": 5.000908851623535,
"learning_rate": 0.00017865515558026428,
"loss": 4.5441,
"step": 35
},
{
"epoch": 0.2822146006859383,
"grad_norm": 6.854843616485596,
"learning_rate": 0.00017698339834299061,
"loss": 4.1148,
"step": 36
},
{
"epoch": 0.2900538951494365,
"grad_norm": 4.968626976013184,
"learning_rate": 0.00017525707698561385,
"loss": 4.2543,
"step": 37
},
{
"epoch": 0.2978931896129348,
"grad_norm": 4.810691833496094,
"learning_rate": 0.00017347741508630672,
"loss": 4.521,
"step": 38
},
{
"epoch": 0.3057324840764331,
"grad_norm": 4.835721015930176,
"learning_rate": 0.00017164567402983152,
"loss": 4.0595,
"step": 39
},
{
"epoch": 0.3135717785399314,
"grad_norm": 4.842018127441406,
"learning_rate": 0.0001697631521134985,
"loss": 4.0795,
"step": 40
},
{
"epoch": 0.3214110730034297,
"grad_norm": 4.61956787109375,
"learning_rate": 0.00016783118362696163,
"loss": 4.0167,
"step": 41
},
{
"epoch": 0.329250367466928,
"grad_norm": 4.931549072265625,
"learning_rate": 0.00016585113790650388,
"loss": 4.1662,
"step": 42
},
{
"epoch": 0.3370896619304263,
"grad_norm": 4.845169544219971,
"learning_rate": 0.00016382441836448202,
"loss": 4.1187,
"step": 43
},
{
"epoch": 0.34492895639392457,
"grad_norm": 4.64799690246582,
"learning_rate": 0.0001617524614946192,
"loss": 4.3107,
"step": 44
},
{
"epoch": 0.3527682508574228,
"grad_norm": 4.621283531188965,
"learning_rate": 0.00015963673585385016,
"loss": 4.2497,
"step": 45
},
{
"epoch": 0.3606075453209211,
"grad_norm": 4.852513313293457,
"learning_rate": 0.0001574787410214407,
"loss": 4.337,
"step": 46
},
{
"epoch": 0.3684468397844194,
"grad_norm": 4.936686038970947,
"learning_rate": 0.00015528000653611935,
"loss": 4.4534,
"step": 47
},
{
"epoch": 0.3762861342479177,
"grad_norm": 4.867508411407471,
"learning_rate": 0.00015304209081197425,
"loss": 3.6969,
"step": 48
},
{
"epoch": 0.384125428711416,
"grad_norm": 4.731996536254883,
"learning_rate": 0.000150766580033884,
"loss": 3.8874,
"step": 49
},
{
"epoch": 0.39196472317491426,
"grad_norm": 4.869089126586914,
"learning_rate": 0.00014845508703326504,
"loss": 5.1876,
"step": 50
},
{
"epoch": 0.39196472317491426,
"eval_loss": 0.2635273039340973,
"eval_runtime": 59.8371,
"eval_samples_per_second": 7.186,
"eval_steps_per_second": 1.805,
"step": 50
},
{
"epoch": 0.39980401763841256,
"grad_norm": 4.403170585632324,
"learning_rate": 0.0001461092501449326,
"loss": 4.2116,
"step": 51
},
{
"epoch": 0.40764331210191085,
"grad_norm": 4.539646625518799,
"learning_rate": 0.00014373073204588556,
"loss": 4.2727,
"step": 52
},
{
"epoch": 0.41548260656540914,
"grad_norm": 4.992217063903809,
"learning_rate": 0.00014132121857683783,
"loss": 4.2613,
"step": 53
},
{
"epoch": 0.4233219010289074,
"grad_norm": 4.952049732208252,
"learning_rate": 0.00013888241754733208,
"loss": 4.3292,
"step": 54
},
{
"epoch": 0.43116119549240567,
"grad_norm": 4.661520481109619,
"learning_rate": 0.00013641605752528224,
"loss": 4.7227,
"step": 55
},
{
"epoch": 0.43900048995590396,
"grad_norm": 4.57968807220459,
"learning_rate": 0.00013392388661180303,
"loss": 4.332,
"step": 56
},
{
"epoch": 0.44683978441940225,
"grad_norm": 4.427809715270996,
"learning_rate": 0.0001314076712021949,
"loss": 4.0129,
"step": 57
},
{
"epoch": 0.45467907888290054,
"grad_norm": 4.379388809204102,
"learning_rate": 0.0001288691947339621,
"loss": 3.625,
"step": 58
},
{
"epoch": 0.46251837334639884,
"grad_norm": 4.387825012207031,
"learning_rate": 0.00012631025642275212,
"loss": 4.0837,
"step": 59
},
{
"epoch": 0.47035766780989713,
"grad_norm": 4.690913677215576,
"learning_rate": 0.0001237326699871115,
"loss": 4.481,
"step": 60
},
{
"epoch": 0.4781969622733954,
"grad_norm": 4.645357608795166,
"learning_rate": 0.00012113826236296244,
"loss": 4.2539,
"step": 61
},
{
"epoch": 0.48603625673689366,
"grad_norm": 4.611278057098389,
"learning_rate": 0.00011852887240871145,
"loss": 4.4024,
"step": 62
},
{
"epoch": 0.49387555120039195,
"grad_norm": 4.584770202636719,
"learning_rate": 0.00011590634960190721,
"loss": 3.708,
"step": 63
},
{
"epoch": 0.5017148456638902,
"grad_norm": 4.215365409851074,
"learning_rate": 0.00011327255272837221,
"loss": 3.9509,
"step": 64
},
{
"epoch": 0.5095541401273885,
"grad_norm": 4.522916316986084,
"learning_rate": 0.00011062934856473655,
"loss": 4.1005,
"step": 65
},
{
"epoch": 0.5173934345908868,
"grad_norm": 4.4469990730285645,
"learning_rate": 0.00010797861055530831,
"loss": 3.5071,
"step": 66
},
{
"epoch": 0.5252327290543851,
"grad_norm": 4.294273376464844,
"learning_rate": 0.00010532221748421787,
"loss": 3.8467,
"step": 67
},
{
"epoch": 0.5330720235178834,
"grad_norm": 4.412168025970459,
"learning_rate": 0.00010266205214377748,
"loss": 4.0123,
"step": 68
},
{
"epoch": 0.5409113179813817,
"grad_norm": 4.205385684967041,
"learning_rate": 0.0001,
"loss": 4.1511,
"step": 69
},
{
"epoch": 0.54875061244488,
"grad_norm": 4.552511692047119,
"learning_rate": 9.733794785622253e-05,
"loss": 3.9935,
"step": 70
},
{
"epoch": 0.5565899069083783,
"grad_norm": 4.413698196411133,
"learning_rate": 9.467778251578217e-05,
"loss": 3.9498,
"step": 71
},
{
"epoch": 0.5644292013718766,
"grad_norm": 4.40337610244751,
"learning_rate": 9.202138944469168e-05,
"loss": 4.3969,
"step": 72
},
{
"epoch": 0.5722684958353749,
"grad_norm": 4.428593158721924,
"learning_rate": 8.937065143526347e-05,
"loss": 3.8083,
"step": 73
},
{
"epoch": 0.580107790298873,
"grad_norm": 4.122357368469238,
"learning_rate": 8.672744727162781e-05,
"loss": 4.0771,
"step": 74
},
{
"epoch": 0.5879470847623713,
"grad_norm": 4.366419792175293,
"learning_rate": 8.409365039809281e-05,
"loss": 4.1221,
"step": 75
},
{
"epoch": 0.5957863792258696,
"grad_norm": 4.299376964569092,
"learning_rate": 8.147112759128859e-05,
"loss": 4.1303,
"step": 76
},
{
"epoch": 0.6036256736893679,
"grad_norm": 4.406579494476318,
"learning_rate": 7.886173763703757e-05,
"loss": 3.9516,
"step": 77
},
{
"epoch": 0.6114649681528662,
"grad_norm": 4.291518211364746,
"learning_rate": 7.626733001288851e-05,
"loss": 4.2969,
"step": 78
},
{
"epoch": 0.6193042626163645,
"grad_norm": 4.067143440246582,
"learning_rate": 7.368974357724789e-05,
"loss": 4.2154,
"step": 79
},
{
"epoch": 0.6271435570798628,
"grad_norm": 4.134098052978516,
"learning_rate": 7.113080526603792e-05,
"loss": 3.7089,
"step": 80
},
{
"epoch": 0.6349828515433611,
"grad_norm": 3.921808958053589,
"learning_rate": 6.859232879780515e-05,
"loss": 4.3355,
"step": 81
},
{
"epoch": 0.6428221460068594,
"grad_norm": 4.351531982421875,
"learning_rate": 6.607611338819697e-05,
"loss": 4.2442,
"step": 82
},
{
"epoch": 0.6506614404703577,
"grad_norm": 4.067043304443359,
"learning_rate": 6.358394247471778e-05,
"loss": 4.1614,
"step": 83
},
{
"epoch": 0.658500734933856,
"grad_norm": 4.547780990600586,
"learning_rate": 6.111758245266794e-05,
"loss": 4.256,
"step": 84
},
{
"epoch": 0.6663400293973543,
"grad_norm": 4.304778575897217,
"learning_rate": 5.867878142316221e-05,
"loss": 4.2114,
"step": 85
},
{
"epoch": 0.6741793238608526,
"grad_norm": 3.7989909648895264,
"learning_rate": 5.626926795411447e-05,
"loss": 3.8491,
"step": 86
},
{
"epoch": 0.6820186183243508,
"grad_norm": 4.27288293838501,
"learning_rate": 5.38907498550674e-05,
"loss": 4.041,
"step": 87
},
{
"epoch": 0.6898579127878491,
"grad_norm": 4.575046062469482,
"learning_rate": 5.1544912966734994e-05,
"loss": 4.0219,
"step": 88
},
{
"epoch": 0.6976972072513474,
"grad_norm": 4.415938854217529,
"learning_rate": 4.9233419966116036e-05,
"loss": 4.0358,
"step": 89
},
{
"epoch": 0.7055365017148456,
"grad_norm": 4.7177934646606445,
"learning_rate": 4.695790918802576e-05,
"loss": 4.0491,
"step": 90
},
{
"epoch": 0.7133757961783439,
"grad_norm": 3.931637763977051,
"learning_rate": 4.47199934638807e-05,
"loss": 3.836,
"step": 91
},
{
"epoch": 0.7212150906418422,
"grad_norm": 4.113886833190918,
"learning_rate": 4.252125897855932e-05,
"loss": 3.6984,
"step": 92
},
{
"epoch": 0.7290543851053405,
"grad_norm": 4.356851100921631,
"learning_rate": 4.036326414614985e-05,
"loss": 4.158,
"step": 93
},
{
"epoch": 0.7368936795688388,
"grad_norm": 4.171023368835449,
"learning_rate": 3.824753850538082e-05,
"loss": 3.8695,
"step": 94
},
{
"epoch": 0.7447329740323371,
"grad_norm": 4.107118606567383,
"learning_rate": 3.617558163551802e-05,
"loss": 3.9032,
"step": 95
},
{
"epoch": 0.7525722684958354,
"grad_norm": 4.645840167999268,
"learning_rate": 3.414886209349615e-05,
"loss": 4.6291,
"step": 96
},
{
"epoch": 0.7604115629593337,
"grad_norm": 3.7909698486328125,
"learning_rate": 3.216881637303839e-05,
"loss": 3.776,
"step": 97
},
{
"epoch": 0.768250857422832,
"grad_norm": 4.037353038787842,
"learning_rate": 3.0236847886501542e-05,
"loss": 3.8556,
"step": 98
},
{
"epoch": 0.7760901518863302,
"grad_norm": 4.654599189758301,
"learning_rate": 2.8354325970168484e-05,
"loss": 3.6748,
"step": 99
},
{
"epoch": 0.7839294463498285,
"grad_norm": 4.140764236450195,
"learning_rate": 2.6522584913693294e-05,
"loss": 3.9498,
"step": 100
},
{
"epoch": 0.7839294463498285,
"eval_loss": 0.24994200468063354,
"eval_runtime": 59.8648,
"eval_samples_per_second": 7.183,
"eval_steps_per_second": 1.804,
"step": 100
},
{
"epoch": 0.7917687408133268,
"grad_norm": 4.1459736824035645,
"learning_rate": 2.4742923014386156e-05,
"loss": 4.0253,
"step": 101
},
{
"epoch": 0.7996080352768251,
"grad_norm": 4.166021347045898,
"learning_rate": 2.301660165700936e-05,
"loss": 4.1193,
"step": 102
},
{
"epoch": 0.8074473297403234,
"grad_norm": 4.5724639892578125,
"learning_rate": 2.1344844419735755e-05,
"loss": 4.1613,
"step": 103
},
{
"epoch": 0.8152866242038217,
"grad_norm": 3.7771105766296387,
"learning_rate": 1.9728836206903656e-05,
"loss": 3.577,
"step": 104
},
{
"epoch": 0.82312591866732,
"grad_norm": 4.427511215209961,
"learning_rate": 1.8169722409183097e-05,
"loss": 4.0308,
"step": 105
},
{
"epoch": 0.8309652131308183,
"grad_norm": 4.253128528594971,
"learning_rate": 1.6668608091748495e-05,
"loss": 4.358,
"step": 106
},
{
"epoch": 0.8388045075943165,
"grad_norm": 3.6124472618103027,
"learning_rate": 1.522655721103291e-05,
"loss": 3.8505,
"step": 107
},
{
"epoch": 0.8466438020578148,
"grad_norm": 4.112203598022461,
"learning_rate": 1.3844591860619383e-05,
"loss": 3.9511,
"step": 108
},
{
"epoch": 0.854483096521313,
"grad_norm": 3.7493703365325928,
"learning_rate": 1.2523691546803873e-05,
"loss": 3.8253,
"step": 109
},
{
"epoch": 0.8623223909848113,
"grad_norm": 4.115346431732178,
"learning_rate": 1.1264792494342857e-05,
"loss": 3.8934,
"step": 110
},
{
"epoch": 0.8701616854483096,
"grad_norm": 4.116607666015625,
"learning_rate": 1.0068786982878087e-05,
"loss": 3.7177,
"step": 111
},
{
"epoch": 0.8780009799118079,
"grad_norm": 3.9183499813079834,
"learning_rate": 8.936522714508678e-06,
"loss": 3.8943,
"step": 112
},
{
"epoch": 0.8858402743753062,
"grad_norm": 4.0384039878845215,
"learning_rate": 7.868802212958703e-06,
"loss": 3.4521,
"step": 113
},
{
"epoch": 0.8936795688388045,
"grad_norm": 4.025205612182617,
"learning_rate": 6.866382254766157e-06,
"loss": 3.8196,
"step": 114
},
{
"epoch": 0.9015188633023028,
"grad_norm": 4.01348352432251,
"learning_rate": 5.929973332896677e-06,
"loss": 3.9242,
"step": 115
},
{
"epoch": 0.9093581577658011,
"grad_norm": 4.146557331085205,
"learning_rate": 5.060239153161872e-06,
"loss": 4.0043,
"step": 116
},
{
"epoch": 0.9171974522292994,
"grad_norm": 4.394860744476318,
"learning_rate": 4.257796163799455e-06,
"loss": 3.8837,
"step": 117
},
{
"epoch": 0.9250367466927977,
"grad_norm": 4.56512451171875,
"learning_rate": 3.5232131185484076e-06,
"loss": 4.3846,
"step": 118
},
{
"epoch": 0.932876041156296,
"grad_norm": 3.989962577819824,
"learning_rate": 2.857010673529015e-06,
"loss": 3.3769,
"step": 119
},
{
"epoch": 0.9407153356197943,
"grad_norm": 4.115790843963623,
"learning_rate": 2.259661018213333e-06,
"loss": 3.7523,
"step": 120
},
{
"epoch": 0.9485546300832925,
"grad_norm": 4.354365348815918,
"learning_rate": 1.7315875407479032e-06,
"loss": 3.7809,
"step": 121
},
{
"epoch": 0.9563939245467908,
"grad_norm": 4.128818035125732,
"learning_rate": 1.2731645278655445e-06,
"loss": 3.8053,
"step": 122
},
{
"epoch": 0.964233219010289,
"grad_norm": 4.223034858703613,
"learning_rate": 8.847168995992916e-07,
"loss": 3.5431,
"step": 123
},
{
"epoch": 0.9720725134737873,
"grad_norm": 4.52358341217041,
"learning_rate": 5.665199789862907e-07,
"loss": 4.4289,
"step": 124
},
{
"epoch": 0.9799118079372856,
"grad_norm": 3.970877170562744,
"learning_rate": 3.1879929692498757e-07,
"loss": 3.9163,
"step": 125
},
{
"epoch": 0.9877511024007839,
"grad_norm": 4.14064359664917,
"learning_rate": 1.4173043232380557e-07,
"loss": 3.9214,
"step": 126
},
{
"epoch": 0.9955903968642822,
"grad_norm": 3.948699712753296,
"learning_rate": 3.5438887654737355e-08,
"loss": 4.2029,
"step": 127
},
{
"epoch": 1.006859382655561,
"grad_norm": 3.799189567565918,
"learning_rate": 0.0,
"loss": 3.387,
"step": 128
}
],
"logging_steps": 1,
"max_steps": 128,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 100,
"stateful_callbacks": {
"EarlyStoppingCallback": {
"args": {
"early_stopping_patience": 2,
"early_stopping_threshold": 0.0
},
"attributes": {
"early_stopping_patience_counter": 0
}
},
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 7.203212659984957e+17,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}