math-vinallama-7b-chat / trainer_state.json
Namronaldo2004's picture
Update fine-tuned model
bb58c6e
raw
history blame
20.7 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 5.0,
"eval_steps": 500,
"global_step": 125,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"grad_norm": 4.460970878601074,
"learning_rate": 2.857142857142857e-05,
"loss": 2.7426,
"step": 1
},
{
"epoch": 0.08,
"grad_norm": 4.278836727142334,
"learning_rate": 5.714285714285714e-05,
"loss": 2.712,
"step": 2
},
{
"epoch": 0.12,
"grad_norm": 3.84088134765625,
"learning_rate": 8.571428571428571e-05,
"loss": 2.546,
"step": 3
},
{
"epoch": 0.16,
"grad_norm": 3.0376698970794678,
"learning_rate": 0.00011428571428571428,
"loss": 2.2798,
"step": 4
},
{
"epoch": 0.2,
"grad_norm": 2.979853391647339,
"learning_rate": 0.00014285714285714287,
"loss": 2.0483,
"step": 5
},
{
"epoch": 0.24,
"grad_norm": 2.246155261993408,
"learning_rate": 0.00017142857142857143,
"loss": 1.7379,
"step": 6
},
{
"epoch": 0.28,
"grad_norm": 1.8112757205963135,
"learning_rate": 0.0002,
"loss": 1.4445,
"step": 7
},
{
"epoch": 0.32,
"grad_norm": 1.477661371231079,
"learning_rate": 0.00019996456111234527,
"loss": 1.2269,
"step": 8
},
{
"epoch": 0.36,
"grad_norm": 1.2332955598831177,
"learning_rate": 0.0001998582695676762,
"loss": 1.0502,
"step": 9
},
{
"epoch": 0.4,
"grad_norm": 1.0987950563430786,
"learning_rate": 0.000199681200703075,
"loss": 0.9511,
"step": 10
},
{
"epoch": 0.44,
"grad_norm": 1.4148105382919312,
"learning_rate": 0.00019943348002101371,
"loss": 0.8553,
"step": 11
},
{
"epoch": 0.48,
"grad_norm": 0.6310116052627563,
"learning_rate": 0.00019911528310040074,
"loss": 0.8427,
"step": 12
},
{
"epoch": 0.52,
"grad_norm": 0.5813988447189331,
"learning_rate": 0.00019872683547213446,
"loss": 0.8571,
"step": 13
},
{
"epoch": 0.56,
"grad_norm": 0.5922828912734985,
"learning_rate": 0.00019826841245925212,
"loss": 0.7995,
"step": 14
},
{
"epoch": 0.6,
"grad_norm": 0.5135756731033325,
"learning_rate": 0.00019774033898178667,
"loss": 0.7967,
"step": 15
},
{
"epoch": 0.64,
"grad_norm": 0.4386445879936218,
"learning_rate": 0.00019714298932647098,
"loss": 0.7655,
"step": 16
},
{
"epoch": 0.68,
"grad_norm": 0.35705333948135376,
"learning_rate": 0.0001964767868814516,
"loss": 0.7255,
"step": 17
},
{
"epoch": 0.72,
"grad_norm": 0.2756179869174957,
"learning_rate": 0.00019574220383620055,
"loss": 0.7296,
"step": 18
},
{
"epoch": 0.76,
"grad_norm": 0.27240365743637085,
"learning_rate": 0.00019493976084683813,
"loss": 0.7523,
"step": 19
},
{
"epoch": 0.8,
"grad_norm": 0.2601045072078705,
"learning_rate": 0.00019407002666710336,
"loss": 0.7477,
"step": 20
},
{
"epoch": 0.84,
"grad_norm": 0.24948769807815552,
"learning_rate": 0.00019313361774523385,
"loss": 0.6847,
"step": 21
},
{
"epoch": 0.88,
"grad_norm": 0.2666375935077667,
"learning_rate": 0.00019213119778704128,
"loss": 0.7327,
"step": 22
},
{
"epoch": 0.92,
"grad_norm": 0.24206262826919556,
"learning_rate": 0.00019106347728549135,
"loss": 0.704,
"step": 23
},
{
"epoch": 0.96,
"grad_norm": 0.2375992089509964,
"learning_rate": 0.00018993121301712193,
"loss": 0.7215,
"step": 24
},
{
"epoch": 1.0,
"grad_norm": 0.2436981201171875,
"learning_rate": 0.00018873520750565718,
"loss": 0.7342,
"step": 25
},
{
"epoch": 1.04,
"grad_norm": 0.23295441269874573,
"learning_rate": 0.00018747630845319612,
"loss": 0.6412,
"step": 26
},
{
"epoch": 1.08,
"grad_norm": 0.23264053463935852,
"learning_rate": 0.0001861554081393806,
"loss": 0.6375,
"step": 27
},
{
"epoch": 1.12,
"grad_norm": 0.24330440163612366,
"learning_rate": 0.0001847734427889671,
"loss": 0.6281,
"step": 28
},
{
"epoch": 1.16,
"grad_norm": 0.226120725274086,
"learning_rate": 0.0001833313919082515,
"loss": 0.6366,
"step": 29
},
{
"epoch": 1.2,
"grad_norm": 0.23621249198913574,
"learning_rate": 0.0001818302775908169,
"loss": 0.6118,
"step": 30
},
{
"epoch": 1.24,
"grad_norm": 0.2544279396533966,
"learning_rate": 0.00018027116379309638,
"loss": 0.6374,
"step": 31
},
{
"epoch": 1.28,
"grad_norm": 0.2579609453678131,
"learning_rate": 0.00017865515558026428,
"loss": 0.6422,
"step": 32
},
{
"epoch": 1.32,
"grad_norm": 0.23667974770069122,
"learning_rate": 0.00017698339834299061,
"loss": 0.6253,
"step": 33
},
{
"epoch": 1.3599999999999999,
"grad_norm": 0.219462051987648,
"learning_rate": 0.00017525707698561385,
"loss": 0.6171,
"step": 34
},
{
"epoch": 1.4,
"grad_norm": 0.22832028567790985,
"learning_rate": 0.00017347741508630672,
"loss": 0.6588,
"step": 35
},
{
"epoch": 1.44,
"grad_norm": 0.2148832529783249,
"learning_rate": 0.00017164567402983152,
"loss": 0.6157,
"step": 36
},
{
"epoch": 1.48,
"grad_norm": 0.22131554782390594,
"learning_rate": 0.0001697631521134985,
"loss": 0.6157,
"step": 37
},
{
"epoch": 1.52,
"grad_norm": 0.21793827414512634,
"learning_rate": 0.00016783118362696163,
"loss": 0.5867,
"step": 38
},
{
"epoch": 1.56,
"grad_norm": 0.2158416360616684,
"learning_rate": 0.00016585113790650388,
"loss": 0.6006,
"step": 39
},
{
"epoch": 1.6,
"grad_norm": 0.2120426446199417,
"learning_rate": 0.00016382441836448202,
"loss": 0.593,
"step": 40
},
{
"epoch": 1.6400000000000001,
"grad_norm": 0.22278062999248505,
"learning_rate": 0.0001617524614946192,
"loss": 0.5768,
"step": 41
},
{
"epoch": 1.6800000000000002,
"grad_norm": 0.21742485463619232,
"learning_rate": 0.00015963673585385016,
"loss": 0.5981,
"step": 42
},
{
"epoch": 1.72,
"grad_norm": 0.22864408791065216,
"learning_rate": 0.0001574787410214407,
"loss": 0.6262,
"step": 43
},
{
"epoch": 1.76,
"grad_norm": 0.2160646617412567,
"learning_rate": 0.00015528000653611935,
"loss": 0.6195,
"step": 44
},
{
"epoch": 1.8,
"grad_norm": 0.21609608829021454,
"learning_rate": 0.00015304209081197425,
"loss": 0.5855,
"step": 45
},
{
"epoch": 1.8399999999999999,
"grad_norm": 0.21653196215629578,
"learning_rate": 0.000150766580033884,
"loss": 0.5852,
"step": 46
},
{
"epoch": 1.88,
"grad_norm": 0.20484659075737,
"learning_rate": 0.00014845508703326504,
"loss": 0.5932,
"step": 47
},
{
"epoch": 1.92,
"grad_norm": 0.214134082198143,
"learning_rate": 0.0001461092501449326,
"loss": 0.5787,
"step": 48
},
{
"epoch": 1.96,
"grad_norm": 0.21110501885414124,
"learning_rate": 0.00014373073204588556,
"loss": 0.572,
"step": 49
},
{
"epoch": 2.0,
"grad_norm": 0.21470192074775696,
"learning_rate": 0.00014132121857683783,
"loss": 0.5655,
"step": 50
},
{
"epoch": 2.04,
"grad_norm": 0.2057720124721527,
"learning_rate": 0.00013888241754733208,
"loss": 0.5304,
"step": 51
},
{
"epoch": 2.08,
"grad_norm": 0.21427400410175323,
"learning_rate": 0.00013641605752528224,
"loss": 0.5319,
"step": 52
},
{
"epoch": 2.12,
"grad_norm": 0.20478470623493195,
"learning_rate": 0.00013392388661180303,
"loss": 0.5296,
"step": 53
},
{
"epoch": 2.16,
"grad_norm": 0.2072669267654419,
"learning_rate": 0.0001314076712021949,
"loss": 0.508,
"step": 54
},
{
"epoch": 2.2,
"grad_norm": 0.22309353947639465,
"learning_rate": 0.0001288691947339621,
"loss": 0.5454,
"step": 55
},
{
"epoch": 2.24,
"grad_norm": 0.22911317646503448,
"learning_rate": 0.00012631025642275212,
"loss": 0.5113,
"step": 56
},
{
"epoch": 2.2800000000000002,
"grad_norm": 0.22047486901283264,
"learning_rate": 0.0001237326699871115,
"loss": 0.5108,
"step": 57
},
{
"epoch": 2.32,
"grad_norm": 0.2026248276233673,
"learning_rate": 0.00012113826236296244,
"loss": 0.4972,
"step": 58
},
{
"epoch": 2.36,
"grad_norm": 0.22186334431171417,
"learning_rate": 0.00011852887240871145,
"loss": 0.473,
"step": 59
},
{
"epoch": 2.4,
"grad_norm": 0.22699356079101562,
"learning_rate": 0.00011590634960190721,
"loss": 0.5094,
"step": 60
},
{
"epoch": 2.44,
"grad_norm": 0.2264937162399292,
"learning_rate": 0.00011327255272837221,
"loss": 0.5157,
"step": 61
},
{
"epoch": 2.48,
"grad_norm": 0.21824264526367188,
"learning_rate": 0.00011062934856473655,
"loss": 0.4982,
"step": 62
},
{
"epoch": 2.52,
"grad_norm": 0.2194606065750122,
"learning_rate": 0.00010797861055530831,
"loss": 0.5322,
"step": 63
},
{
"epoch": 2.56,
"grad_norm": 0.24011456966400146,
"learning_rate": 0.00010532221748421787,
"loss": 0.5426,
"step": 64
},
{
"epoch": 2.6,
"grad_norm": 0.23064349591732025,
"learning_rate": 0.00010266205214377748,
"loss": 0.5237,
"step": 65
},
{
"epoch": 2.64,
"grad_norm": 0.23363706469535828,
"learning_rate": 0.0001,
"loss": 0.504,
"step": 66
},
{
"epoch": 2.68,
"grad_norm": 0.22657082974910736,
"learning_rate": 9.733794785622253e-05,
"loss": 0.4921,
"step": 67
},
{
"epoch": 2.7199999999999998,
"grad_norm": 0.2323610782623291,
"learning_rate": 9.467778251578217e-05,
"loss": 0.5026,
"step": 68
},
{
"epoch": 2.76,
"grad_norm": 0.22157961130142212,
"learning_rate": 9.202138944469168e-05,
"loss": 0.4814,
"step": 69
},
{
"epoch": 2.8,
"grad_norm": 0.2387864887714386,
"learning_rate": 8.937065143526347e-05,
"loss": 0.512,
"step": 70
},
{
"epoch": 2.84,
"grad_norm": 0.22291497886180878,
"learning_rate": 8.672744727162781e-05,
"loss": 0.5037,
"step": 71
},
{
"epoch": 2.88,
"grad_norm": 0.22832155227661133,
"learning_rate": 8.409365039809281e-05,
"loss": 0.4887,
"step": 72
},
{
"epoch": 2.92,
"grad_norm": 0.22989900410175323,
"learning_rate": 8.147112759128859e-05,
"loss": 0.5223,
"step": 73
},
{
"epoch": 2.96,
"grad_norm": 0.23885484039783478,
"learning_rate": 7.886173763703757e-05,
"loss": 0.5121,
"step": 74
},
{
"epoch": 3.0,
"grad_norm": 0.24740351736545563,
"learning_rate": 7.626733001288851e-05,
"loss": 0.5035,
"step": 75
},
{
"epoch": 3.04,
"grad_norm": 0.22852157056331635,
"learning_rate": 7.368974357724789e-05,
"loss": 0.4772,
"step": 76
},
{
"epoch": 3.08,
"grad_norm": 0.24161659181118011,
"learning_rate": 7.113080526603792e-05,
"loss": 0.4723,
"step": 77
},
{
"epoch": 3.12,
"grad_norm": 0.23222312331199646,
"learning_rate": 6.859232879780515e-05,
"loss": 0.4573,
"step": 78
},
{
"epoch": 3.16,
"grad_norm": 0.2266985923051834,
"learning_rate": 6.607611338819697e-05,
"loss": 0.4845,
"step": 79
},
{
"epoch": 3.2,
"grad_norm": 0.2399943470954895,
"learning_rate": 6.358394247471778e-05,
"loss": 0.4522,
"step": 80
},
{
"epoch": 3.24,
"grad_norm": 0.24266791343688965,
"learning_rate": 6.111758245266794e-05,
"loss": 0.4546,
"step": 81
},
{
"epoch": 3.2800000000000002,
"grad_norm": 0.2595340609550476,
"learning_rate": 5.867878142316221e-05,
"loss": 0.4648,
"step": 82
},
{
"epoch": 3.32,
"grad_norm": 0.23132656514644623,
"learning_rate": 5.626926795411447e-05,
"loss": 0.418,
"step": 83
},
{
"epoch": 3.36,
"grad_norm": 0.2526528239250183,
"learning_rate": 5.38907498550674e-05,
"loss": 0.4485,
"step": 84
},
{
"epoch": 3.4,
"grad_norm": 0.24199239909648895,
"learning_rate": 5.1544912966734994e-05,
"loss": 0.4466,
"step": 85
},
{
"epoch": 3.44,
"grad_norm": 0.24821186065673828,
"learning_rate": 4.9233419966116036e-05,
"loss": 0.4471,
"step": 86
},
{
"epoch": 3.48,
"grad_norm": 0.2661694586277008,
"learning_rate": 4.695790918802576e-05,
"loss": 0.4555,
"step": 87
},
{
"epoch": 3.52,
"grad_norm": 0.24729236960411072,
"learning_rate": 4.47199934638807e-05,
"loss": 0.4459,
"step": 88
},
{
"epoch": 3.56,
"grad_norm": 0.24867290258407593,
"learning_rate": 4.252125897855932e-05,
"loss": 0.436,
"step": 89
},
{
"epoch": 3.6,
"grad_norm": 0.24836455285549164,
"learning_rate": 4.036326414614985e-05,
"loss": 0.433,
"step": 90
},
{
"epoch": 3.64,
"grad_norm": 0.2629256844520569,
"learning_rate": 3.824753850538082e-05,
"loss": 0.4329,
"step": 91
},
{
"epoch": 3.68,
"grad_norm": 0.25363099575042725,
"learning_rate": 3.617558163551802e-05,
"loss": 0.4162,
"step": 92
},
{
"epoch": 3.7199999999999998,
"grad_norm": 0.256331205368042,
"learning_rate": 3.414886209349615e-05,
"loss": 0.4352,
"step": 93
},
{
"epoch": 3.76,
"grad_norm": 0.259033203125,
"learning_rate": 3.216881637303839e-05,
"loss": 0.4539,
"step": 94
},
{
"epoch": 3.8,
"grad_norm": 0.2493227869272232,
"learning_rate": 3.0236847886501542e-05,
"loss": 0.4232,
"step": 95
},
{
"epoch": 3.84,
"grad_norm": 0.2529692053794861,
"learning_rate": 2.8354325970168484e-05,
"loss": 0.4448,
"step": 96
},
{
"epoch": 3.88,
"grad_norm": 0.25735777616500854,
"learning_rate": 2.6522584913693294e-05,
"loss": 0.4428,
"step": 97
},
{
"epoch": 3.92,
"grad_norm": 0.25182002782821655,
"learning_rate": 2.4742923014386156e-05,
"loss": 0.4126,
"step": 98
},
{
"epoch": 3.96,
"grad_norm": 0.2627415657043457,
"learning_rate": 2.301660165700936e-05,
"loss": 0.4499,
"step": 99
},
{
"epoch": 4.0,
"grad_norm": 0.2550097703933716,
"learning_rate": 2.1344844419735755e-05,
"loss": 0.4174,
"step": 100
},
{
"epoch": 4.04,
"grad_norm": 0.2350243628025055,
"learning_rate": 1.9728836206903656e-05,
"loss": 0.4139,
"step": 101
},
{
"epoch": 4.08,
"grad_norm": 0.24459272623062134,
"learning_rate": 1.8169722409183097e-05,
"loss": 0.3973,
"step": 102
},
{
"epoch": 4.12,
"grad_norm": 0.2467670738697052,
"learning_rate": 1.6668608091748495e-05,
"loss": 0.4252,
"step": 103
},
{
"epoch": 4.16,
"grad_norm": 0.24727074801921844,
"learning_rate": 1.522655721103291e-05,
"loss": 0.3919,
"step": 104
},
{
"epoch": 4.2,
"grad_norm": 0.24538907408714294,
"learning_rate": 1.3844591860619383e-05,
"loss": 0.4517,
"step": 105
},
{
"epoch": 4.24,
"grad_norm": 0.2430519461631775,
"learning_rate": 1.2523691546803873e-05,
"loss": 0.4272,
"step": 106
},
{
"epoch": 4.28,
"grad_norm": 0.2513393759727478,
"learning_rate": 1.1264792494342857e-05,
"loss": 0.4022,
"step": 107
},
{
"epoch": 4.32,
"grad_norm": 0.24626919627189636,
"learning_rate": 1.0068786982878087e-05,
"loss": 0.405,
"step": 108
},
{
"epoch": 4.36,
"grad_norm": 0.25008630752563477,
"learning_rate": 8.936522714508678e-06,
"loss": 0.4123,
"step": 109
},
{
"epoch": 4.4,
"grad_norm": 0.2524281442165375,
"learning_rate": 7.868802212958703e-06,
"loss": 0.428,
"step": 110
},
{
"epoch": 4.44,
"grad_norm": 0.24142776429653168,
"learning_rate": 6.866382254766157e-06,
"loss": 0.4291,
"step": 111
},
{
"epoch": 4.48,
"grad_norm": 0.24990609288215637,
"learning_rate": 5.929973332896677e-06,
"loss": 0.3961,
"step": 112
},
{
"epoch": 4.52,
"grad_norm": 0.23870083689689636,
"learning_rate": 5.060239153161872e-06,
"loss": 0.4001,
"step": 113
},
{
"epoch": 4.5600000000000005,
"grad_norm": 0.2507989704608917,
"learning_rate": 4.257796163799455e-06,
"loss": 0.386,
"step": 114
},
{
"epoch": 4.6,
"grad_norm": 0.244070902466774,
"learning_rate": 3.5232131185484076e-06,
"loss": 0.4357,
"step": 115
},
{
"epoch": 4.64,
"grad_norm": 0.24638234078884125,
"learning_rate": 2.857010673529015e-06,
"loss": 0.3986,
"step": 116
},
{
"epoch": 4.68,
"grad_norm": 0.24885112047195435,
"learning_rate": 2.259661018213333e-06,
"loss": 0.3969,
"step": 117
},
{
"epoch": 4.72,
"grad_norm": 0.24754343926906586,
"learning_rate": 1.7315875407479032e-06,
"loss": 0.3918,
"step": 118
},
{
"epoch": 4.76,
"grad_norm": 0.2457382082939148,
"learning_rate": 1.2731645278655445e-06,
"loss": 0.3882,
"step": 119
},
{
"epoch": 4.8,
"grad_norm": 0.2584328353404999,
"learning_rate": 8.847168995992916e-07,
"loss": 0.419,
"step": 120
},
{
"epoch": 4.84,
"grad_norm": 0.24872322380542755,
"learning_rate": 5.665199789862907e-07,
"loss": 0.4084,
"step": 121
},
{
"epoch": 4.88,
"grad_norm": 0.26830849051475525,
"learning_rate": 3.1879929692498757e-07,
"loss": 0.4304,
"step": 122
},
{
"epoch": 4.92,
"grad_norm": 0.24320735037326813,
"learning_rate": 1.4173043232380557e-07,
"loss": 0.4183,
"step": 123
},
{
"epoch": 4.96,
"grad_norm": 0.272348552942276,
"learning_rate": 3.5438887654737355e-08,
"loss": 0.4105,
"step": 124
},
{
"epoch": 5.0,
"grad_norm": 0.24668046832084656,
"learning_rate": 0.0,
"loss": 0.4111,
"step": 125
}
],
"logging_steps": 1,
"max_steps": 125,
"num_input_tokens_seen": 0,
"num_train_epochs": 5,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.5177777819849523e+17,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}