ncbateman's picture
Training in progress, step 105, checkpoint
ccc3ef1 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0023866348448687,
"eval_steps": 27,
"global_step": 105,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.00954653937947494,
"grad_norm": 0.5416718125343323,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.8067,
"step": 1
},
{
"epoch": 0.00954653937947494,
"eval_loss": 0.8745781183242798,
"eval_runtime": 23.7103,
"eval_samples_per_second": 7.465,
"eval_steps_per_second": 0.97,
"step": 1
},
{
"epoch": 0.01909307875894988,
"grad_norm": 0.34453752636909485,
"learning_rate": 4.000000000000001e-06,
"loss": 0.8173,
"step": 2
},
{
"epoch": 0.028639618138424822,
"grad_norm": 0.4350746273994446,
"learning_rate": 6e-06,
"loss": 0.8136,
"step": 3
},
{
"epoch": 0.03818615751789976,
"grad_norm": 0.32216551899909973,
"learning_rate": 8.000000000000001e-06,
"loss": 0.8387,
"step": 4
},
{
"epoch": 0.0477326968973747,
"grad_norm": 0.6193002462387085,
"learning_rate": 1e-05,
"loss": 0.8368,
"step": 5
},
{
"epoch": 0.057279236276849645,
"grad_norm": 0.3906833529472351,
"learning_rate": 1.2e-05,
"loss": 0.8408,
"step": 6
},
{
"epoch": 0.06682577565632458,
"grad_norm": 0.43147146701812744,
"learning_rate": 1.4000000000000001e-05,
"loss": 0.8195,
"step": 7
},
{
"epoch": 0.07637231503579953,
"grad_norm": 0.300281286239624,
"learning_rate": 1.6000000000000003e-05,
"loss": 0.8286,
"step": 8
},
{
"epoch": 0.08591885441527446,
"grad_norm": 0.36642926931381226,
"learning_rate": 1.8e-05,
"loss": 0.9017,
"step": 9
},
{
"epoch": 0.0954653937947494,
"grad_norm": 0.39945727586746216,
"learning_rate": 2e-05,
"loss": 1.0405,
"step": 10
},
{
"epoch": 0.10501193317422435,
"grad_norm": 0.4177384078502655,
"learning_rate": 2.2000000000000003e-05,
"loss": 0.7997,
"step": 11
},
{
"epoch": 0.11455847255369929,
"grad_norm": 0.6147989630699158,
"learning_rate": 2.4e-05,
"loss": 0.8825,
"step": 12
},
{
"epoch": 0.12410501193317422,
"grad_norm": 0.29250773787498474,
"learning_rate": 2.6000000000000002e-05,
"loss": 0.8231,
"step": 13
},
{
"epoch": 0.13365155131264916,
"grad_norm": 0.4127359092235565,
"learning_rate": 2.8000000000000003e-05,
"loss": 0.8333,
"step": 14
},
{
"epoch": 0.1431980906921241,
"grad_norm": 0.3746739327907562,
"learning_rate": 3e-05,
"loss": 0.9095,
"step": 15
},
{
"epoch": 0.15274463007159905,
"grad_norm": 0.4909857213497162,
"learning_rate": 3.2000000000000005e-05,
"loss": 0.9133,
"step": 16
},
{
"epoch": 0.162291169451074,
"grad_norm": 0.357906699180603,
"learning_rate": 3.4000000000000007e-05,
"loss": 0.8572,
"step": 17
},
{
"epoch": 0.1718377088305489,
"grad_norm": 0.533477783203125,
"learning_rate": 3.6e-05,
"loss": 0.9061,
"step": 18
},
{
"epoch": 0.18138424821002386,
"grad_norm": 0.5812142491340637,
"learning_rate": 3.8e-05,
"loss": 0.7458,
"step": 19
},
{
"epoch": 0.1909307875894988,
"grad_norm": 0.7447901964187622,
"learning_rate": 4e-05,
"loss": 0.7656,
"step": 20
},
{
"epoch": 0.20047732696897375,
"grad_norm": 0.5314919948577881,
"learning_rate": 4.2e-05,
"loss": 0.7249,
"step": 21
},
{
"epoch": 0.2100238663484487,
"grad_norm": 0.6344147324562073,
"learning_rate": 4.4000000000000006e-05,
"loss": 0.7594,
"step": 22
},
{
"epoch": 0.21957040572792363,
"grad_norm": 0.3820248246192932,
"learning_rate": 4.600000000000001e-05,
"loss": 0.7272,
"step": 23
},
{
"epoch": 0.22911694510739858,
"grad_norm": 0.47990524768829346,
"learning_rate": 4.8e-05,
"loss": 0.7684,
"step": 24
},
{
"epoch": 0.2386634844868735,
"grad_norm": 0.5735926032066345,
"learning_rate": 5e-05,
"loss": 0.7445,
"step": 25
},
{
"epoch": 0.24821002386634844,
"grad_norm": 0.3894128203392029,
"learning_rate": 5.2000000000000004e-05,
"loss": 0.8173,
"step": 26
},
{
"epoch": 0.2577565632458234,
"grad_norm": 0.46286776661872864,
"learning_rate": 5.4000000000000005e-05,
"loss": 0.7138,
"step": 27
},
{
"epoch": 0.2577565632458234,
"eval_loss": 0.7636813521385193,
"eval_runtime": 23.8391,
"eval_samples_per_second": 7.425,
"eval_steps_per_second": 0.965,
"step": 27
},
{
"epoch": 0.26730310262529833,
"grad_norm": 0.35924747586250305,
"learning_rate": 5.6000000000000006e-05,
"loss": 0.7682,
"step": 28
},
{
"epoch": 0.27684964200477324,
"grad_norm": 0.8549047708511353,
"learning_rate": 5.8e-05,
"loss": 0.7115,
"step": 29
},
{
"epoch": 0.2863961813842482,
"grad_norm": 0.39896008372306824,
"learning_rate": 6e-05,
"loss": 0.8621,
"step": 30
},
{
"epoch": 0.29594272076372313,
"grad_norm": 0.4116915464401245,
"learning_rate": 6.2e-05,
"loss": 0.682,
"step": 31
},
{
"epoch": 0.3054892601431981,
"grad_norm": 0.33658358454704285,
"learning_rate": 6.400000000000001e-05,
"loss": 0.7713,
"step": 32
},
{
"epoch": 0.315035799522673,
"grad_norm": 0.7733840346336365,
"learning_rate": 6.6e-05,
"loss": 0.683,
"step": 33
},
{
"epoch": 0.324582338902148,
"grad_norm": 0.38960912823677063,
"learning_rate": 6.800000000000001e-05,
"loss": 0.6601,
"step": 34
},
{
"epoch": 0.3341288782816229,
"grad_norm": 0.5905539989471436,
"learning_rate": 7e-05,
"loss": 0.7086,
"step": 35
},
{
"epoch": 0.3436754176610978,
"grad_norm": 0.5276817083358765,
"learning_rate": 7.2e-05,
"loss": 0.7206,
"step": 36
},
{
"epoch": 0.3532219570405728,
"grad_norm": 0.5476186275482178,
"learning_rate": 7.4e-05,
"loss": 0.7032,
"step": 37
},
{
"epoch": 0.3627684964200477,
"grad_norm": 0.5326115489006042,
"learning_rate": 7.6e-05,
"loss": 0.734,
"step": 38
},
{
"epoch": 0.3723150357995227,
"grad_norm": 0.3078496754169464,
"learning_rate": 7.800000000000001e-05,
"loss": 0.7866,
"step": 39
},
{
"epoch": 0.3818615751789976,
"grad_norm": 0.269064724445343,
"learning_rate": 8e-05,
"loss": 0.7215,
"step": 40
},
{
"epoch": 0.3914081145584726,
"grad_norm": 0.30524760484695435,
"learning_rate": 8.2e-05,
"loss": 0.7716,
"step": 41
},
{
"epoch": 0.4009546539379475,
"grad_norm": 0.29701563715934753,
"learning_rate": 8.4e-05,
"loss": 0.7049,
"step": 42
},
{
"epoch": 0.4105011933174224,
"grad_norm": 0.30442681908607483,
"learning_rate": 8.6e-05,
"loss": 0.7702,
"step": 43
},
{
"epoch": 0.4200477326968974,
"grad_norm": 0.2677440345287323,
"learning_rate": 8.800000000000001e-05,
"loss": 0.6415,
"step": 44
},
{
"epoch": 0.4295942720763723,
"grad_norm": 0.4278031289577484,
"learning_rate": 9e-05,
"loss": 0.681,
"step": 45
},
{
"epoch": 0.43914081145584727,
"grad_norm": 0.30344948172569275,
"learning_rate": 9.200000000000001e-05,
"loss": 0.7682,
"step": 46
},
{
"epoch": 0.4486873508353222,
"grad_norm": 0.28526848554611206,
"learning_rate": 9.4e-05,
"loss": 0.6851,
"step": 47
},
{
"epoch": 0.45823389021479716,
"grad_norm": 0.2424936592578888,
"learning_rate": 9.6e-05,
"loss": 0.6337,
"step": 48
},
{
"epoch": 0.4677804295942721,
"grad_norm": 0.44285890460014343,
"learning_rate": 9.8e-05,
"loss": 0.7426,
"step": 49
},
{
"epoch": 0.477326968973747,
"grad_norm": 0.7494658827781677,
"learning_rate": 0.0001,
"loss": 0.7526,
"step": 50
},
{
"epoch": 0.48687350835322196,
"grad_norm": 0.359139621257782,
"learning_rate": 9.991845519630678e-05,
"loss": 0.7343,
"step": 51
},
{
"epoch": 0.4964200477326969,
"grad_norm": 0.3556593358516693,
"learning_rate": 9.967408676742751e-05,
"loss": 0.7711,
"step": 52
},
{
"epoch": 0.5059665871121718,
"grad_norm": 0.3111443817615509,
"learning_rate": 9.926769179238466e-05,
"loss": 0.7054,
"step": 53
},
{
"epoch": 0.5155131264916468,
"grad_norm": 0.3471980690956116,
"learning_rate": 9.870059584711668e-05,
"loss": 0.7179,
"step": 54
},
{
"epoch": 0.5155131264916468,
"eval_loss": 0.7270187139511108,
"eval_runtime": 23.8463,
"eval_samples_per_second": 7.423,
"eval_steps_per_second": 0.965,
"step": 54
},
{
"epoch": 0.5250596658711217,
"grad_norm": 0.3021623492240906,
"learning_rate": 9.797464868072488e-05,
"loss": 0.7227,
"step": 55
},
{
"epoch": 0.5346062052505967,
"grad_norm": 0.391401082277298,
"learning_rate": 9.709221818197624e-05,
"loss": 0.6911,
"step": 56
},
{
"epoch": 0.5441527446300716,
"grad_norm": 0.6585782170295715,
"learning_rate": 9.60561826557425e-05,
"loss": 0.6337,
"step": 57
},
{
"epoch": 0.5536992840095465,
"grad_norm": 0.35819312930107117,
"learning_rate": 9.486992143456792e-05,
"loss": 0.8028,
"step": 58
},
{
"epoch": 0.5632458233890215,
"grad_norm": 0.34819847345352173,
"learning_rate": 9.353730385598887e-05,
"loss": 0.6767,
"step": 59
},
{
"epoch": 0.5727923627684964,
"grad_norm": 0.2741661071777344,
"learning_rate": 9.206267664155907e-05,
"loss": 0.6573,
"step": 60
},
{
"epoch": 0.5823389021479713,
"grad_norm": 0.37481749057769775,
"learning_rate": 9.045084971874738e-05,
"loss": 0.7067,
"step": 61
},
{
"epoch": 0.5918854415274463,
"grad_norm": 0.27204594016075134,
"learning_rate": 8.870708053195413e-05,
"loss": 0.6636,
"step": 62
},
{
"epoch": 0.6014319809069213,
"grad_norm": 3.2921030521392822,
"learning_rate": 8.683705689382024e-05,
"loss": 0.7162,
"step": 63
},
{
"epoch": 0.6109785202863962,
"grad_norm": 0.34250608086586,
"learning_rate": 8.484687843276469e-05,
"loss": 0.7088,
"step": 64
},
{
"epoch": 0.6205250596658711,
"grad_norm": 0.412442147731781,
"learning_rate": 8.274303669726426e-05,
"loss": 0.6562,
"step": 65
},
{
"epoch": 0.630071599045346,
"grad_norm": 0.2742518186569214,
"learning_rate": 8.053239398177191e-05,
"loss": 0.7462,
"step": 66
},
{
"epoch": 0.639618138424821,
"grad_norm": 0.28374218940734863,
"learning_rate": 7.822216094333847e-05,
"loss": 0.727,
"step": 67
},
{
"epoch": 0.649164677804296,
"grad_norm": 0.36977893114089966,
"learning_rate": 7.58198730819481e-05,
"loss": 0.7014,
"step": 68
},
{
"epoch": 0.6587112171837709,
"grad_norm": 0.3121008574962616,
"learning_rate": 7.333336616128369e-05,
"loss": 0.7509,
"step": 69
},
{
"epoch": 0.6682577565632458,
"grad_norm": 0.3306713402271271,
"learning_rate": 7.077075065009433e-05,
"loss": 0.7309,
"step": 70
},
{
"epoch": 0.6778042959427207,
"grad_norm": 0.4980546832084656,
"learning_rate": 6.814038526753205e-05,
"loss": 0.7188,
"step": 71
},
{
"epoch": 0.6873508353221957,
"grad_norm": 0.36510801315307617,
"learning_rate": 6.545084971874738e-05,
"loss": 0.7224,
"step": 72
},
{
"epoch": 0.6968973747016707,
"grad_norm": 0.3253666162490845,
"learning_rate": 6.271091670967436e-05,
"loss": 0.7039,
"step": 73
},
{
"epoch": 0.7064439140811456,
"grad_norm": 0.46175265312194824,
"learning_rate": 5.992952333228728e-05,
"loss": 0.7333,
"step": 74
},
{
"epoch": 0.7159904534606205,
"grad_norm": 0.27166417241096497,
"learning_rate": 5.7115741913664264e-05,
"loss": 0.7273,
"step": 75
},
{
"epoch": 0.7255369928400954,
"grad_norm": 0.25115492939949036,
"learning_rate": 5.427875042394199e-05,
"loss": 0.7164,
"step": 76
},
{
"epoch": 0.7350835322195705,
"grad_norm": 0.3494699001312256,
"learning_rate": 5.142780253968481e-05,
"loss": 0.8032,
"step": 77
},
{
"epoch": 0.7446300715990454,
"grad_norm": 0.31800737977027893,
"learning_rate": 4.85721974603152e-05,
"loss": 0.7018,
"step": 78
},
{
"epoch": 0.7541766109785203,
"grad_norm": 0.44433706998825073,
"learning_rate": 4.5721249576058027e-05,
"loss": 0.7192,
"step": 79
},
{
"epoch": 0.7637231503579952,
"grad_norm": 0.33603635430336,
"learning_rate": 4.288425808633575e-05,
"loss": 0.66,
"step": 80
},
{
"epoch": 0.7732696897374701,
"grad_norm": 0.25413641333580017,
"learning_rate": 4.007047666771274e-05,
"loss": 0.5788,
"step": 81
},
{
"epoch": 0.7732696897374701,
"eval_loss": 0.7156311273574829,
"eval_runtime": 24.2989,
"eval_samples_per_second": 7.284,
"eval_steps_per_second": 0.947,
"step": 81
},
{
"epoch": 0.7828162291169452,
"grad_norm": 0.5405492782592773,
"learning_rate": 3.728908329032567e-05,
"loss": 0.7513,
"step": 82
},
{
"epoch": 0.7923627684964201,
"grad_norm": 0.33515989780426025,
"learning_rate": 3.4549150281252636e-05,
"loss": 0.6574,
"step": 83
},
{
"epoch": 0.801909307875895,
"grad_norm": 0.3084225058555603,
"learning_rate": 3.1859614732467954e-05,
"loss": 0.6975,
"step": 84
},
{
"epoch": 0.8114558472553699,
"grad_norm": 0.3069574534893036,
"learning_rate": 2.9229249349905684e-05,
"loss": 0.6461,
"step": 85
},
{
"epoch": 0.8210023866348448,
"grad_norm": 1.0965303182601929,
"learning_rate": 2.6666633838716314e-05,
"loss": 0.6629,
"step": 86
},
{
"epoch": 0.8305489260143198,
"grad_norm": 0.36803197860717773,
"learning_rate": 2.418012691805191e-05,
"loss": 0.7792,
"step": 87
},
{
"epoch": 0.8400954653937948,
"grad_norm": 0.37279656529426575,
"learning_rate": 2.1777839056661554e-05,
"loss": 0.7039,
"step": 88
},
{
"epoch": 0.8496420047732697,
"grad_norm": 0.28656861186027527,
"learning_rate": 1.946760601822809e-05,
"loss": 0.7103,
"step": 89
},
{
"epoch": 0.8591885441527446,
"grad_norm": 0.805375874042511,
"learning_rate": 1.725696330273575e-05,
"loss": 0.7213,
"step": 90
},
{
"epoch": 0.8687350835322196,
"grad_norm": 0.3051246404647827,
"learning_rate": 1.5153121567235335e-05,
"loss": 0.6856,
"step": 91
},
{
"epoch": 0.8782816229116945,
"grad_norm": 0.6019058227539062,
"learning_rate": 1.3162943106179749e-05,
"loss": 0.7098,
"step": 92
},
{
"epoch": 0.8878281622911695,
"grad_norm": 0.3393521308898926,
"learning_rate": 1.1292919468045877e-05,
"loss": 0.8349,
"step": 93
},
{
"epoch": 0.8973747016706444,
"grad_norm": 0.2818894386291504,
"learning_rate": 9.549150281252633e-06,
"loss": 0.7454,
"step": 94
},
{
"epoch": 0.9069212410501193,
"grad_norm": 0.295830637216568,
"learning_rate": 7.937323358440935e-06,
"loss": 0.796,
"step": 95
},
{
"epoch": 0.9164677804295943,
"grad_norm": 0.3201135993003845,
"learning_rate": 6.462696144011149e-06,
"loss": 0.7365,
"step": 96
},
{
"epoch": 0.9260143198090692,
"grad_norm": 0.481950968503952,
"learning_rate": 5.13007856543209e-06,
"loss": 0.6355,
"step": 97
},
{
"epoch": 0.9355608591885441,
"grad_norm": 0.45520639419555664,
"learning_rate": 3.9438173442575e-06,
"loss": 0.6937,
"step": 98
},
{
"epoch": 0.9451073985680191,
"grad_norm": 0.6622302532196045,
"learning_rate": 2.9077818180237693e-06,
"loss": 0.6094,
"step": 99
},
{
"epoch": 0.954653937947494,
"grad_norm": 0.4538000822067261,
"learning_rate": 2.0253513192751373e-06,
"loss": 0.6478,
"step": 100
},
{
"epoch": 0.964200477326969,
"grad_norm": 0.3268328607082367,
"learning_rate": 1.2994041528833266e-06,
"loss": 0.6785,
"step": 101
},
{
"epoch": 0.9737470167064439,
"grad_norm": 0.24223873019218445,
"learning_rate": 7.323082076153509e-07,
"loss": 0.6202,
"step": 102
},
{
"epoch": 0.9832935560859188,
"grad_norm": 0.38341718912124634,
"learning_rate": 3.2591323257248893e-07,
"loss": 0.7639,
"step": 103
},
{
"epoch": 0.9928400954653938,
"grad_norm": 0.32395732402801514,
"learning_rate": 8.15448036932176e-08,
"loss": 0.6257,
"step": 104
},
{
"epoch": 1.0023866348448687,
"grad_norm": 0.4013065993785858,
"learning_rate": 0.0,
"loss": 0.6666,
"step": 105
}
],
"logging_steps": 1,
"max_steps": 105,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 6.231853181907763e+17,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}