dada22231's picture
Training in progress, step 95, checkpoint
a06546f verified
{
"best_metric": 1.7268242835998535,
"best_model_checkpoint": "miner_id_24/checkpoint-75",
"epoch": 0.628358825961141,
"eval_steps": 25,
"global_step": 95,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.006614303431169905,
"grad_norm": 17.736679077148438,
"learning_rate": 3.3333333333333335e-05,
"loss": 11.468,
"step": 1
},
{
"epoch": 0.006614303431169905,
"eval_loss": 11.592829704284668,
"eval_runtime": 4.0134,
"eval_samples_per_second": 12.458,
"eval_steps_per_second": 3.239,
"step": 1
},
{
"epoch": 0.01322860686233981,
"grad_norm": 20.307588577270508,
"learning_rate": 6.666666666666667e-05,
"loss": 11.3596,
"step": 2
},
{
"epoch": 0.019842910293509715,
"grad_norm": 15.574076652526855,
"learning_rate": 0.0001,
"loss": 10.8161,
"step": 3
},
{
"epoch": 0.02645721372467962,
"grad_norm": 12.424201011657715,
"learning_rate": 9.997376600647783e-05,
"loss": 9.4755,
"step": 4
},
{
"epoch": 0.03307151715584952,
"grad_norm": 9.448874473571777,
"learning_rate": 9.989509461357426e-05,
"loss": 8.0971,
"step": 5
},
{
"epoch": 0.03968582058701943,
"grad_norm": 10.42374038696289,
"learning_rate": 9.976407754861426e-05,
"loss": 7.2079,
"step": 6
},
{
"epoch": 0.04630012401818934,
"grad_norm": 12.758567810058594,
"learning_rate": 9.958086757163489e-05,
"loss": 5.7105,
"step": 7
},
{
"epoch": 0.05291442744935924,
"grad_norm": 13.52912712097168,
"learning_rate": 9.934567829727386e-05,
"loss": 3.8279,
"step": 8
},
{
"epoch": 0.059528730880529145,
"grad_norm": 15.979461669921875,
"learning_rate": 9.905878394570453e-05,
"loss": 3.0876,
"step": 9
},
{
"epoch": 0.06614303431169904,
"grad_norm": 7.501053333282471,
"learning_rate": 9.872051902290737e-05,
"loss": 2.8203,
"step": 10
},
{
"epoch": 0.07275733774286895,
"grad_norm": 7.698332786560059,
"learning_rate": 9.833127793065098e-05,
"loss": 3.329,
"step": 11
},
{
"epoch": 0.07937164117403886,
"grad_norm": 7.099416732788086,
"learning_rate": 9.789151450663723e-05,
"loss": 2.6664,
"step": 12
},
{
"epoch": 0.08598594460520877,
"grad_norm": 15.326375961303711,
"learning_rate": 9.740174149534693e-05,
"loss": 2.7908,
"step": 13
},
{
"epoch": 0.09260024803637867,
"grad_norm": 10.195456504821777,
"learning_rate": 9.686252995020249e-05,
"loss": 2.3301,
"step": 14
},
{
"epoch": 0.09921455146754857,
"grad_norm": 4.361770153045654,
"learning_rate": 9.627450856774539e-05,
"loss": 2.0583,
"step": 15
},
{
"epoch": 0.10582885489871847,
"grad_norm": 3.182706117630005,
"learning_rate": 9.563836295460398e-05,
"loss": 2.0367,
"step": 16
},
{
"epoch": 0.11244315832988838,
"grad_norm": 5.076944828033447,
"learning_rate": 9.495483482810688e-05,
"loss": 1.9523,
"step": 17
},
{
"epoch": 0.11905746176105829,
"grad_norm": 8.57449722290039,
"learning_rate": 9.422472115147382e-05,
"loss": 2.025,
"step": 18
},
{
"epoch": 0.1256717651922282,
"grad_norm": 3.4628255367279053,
"learning_rate": 9.3448873204592e-05,
"loss": 2.0057,
"step": 19
},
{
"epoch": 0.1322860686233981,
"grad_norm": 2.956812620162964,
"learning_rate": 9.2628195591462e-05,
"loss": 2.006,
"step": 20
},
{
"epoch": 0.138900372054568,
"grad_norm": 9.580174446105957,
"learning_rate": 9.176364518546989e-05,
"loss": 2.0139,
"step": 21
},
{
"epoch": 0.1455146754857379,
"grad_norm": 6.208186626434326,
"learning_rate": 9.08562300137157e-05,
"loss": 1.9206,
"step": 22
},
{
"epoch": 0.15212897891690783,
"grad_norm": 4.192629337310791,
"learning_rate": 8.990700808169889e-05,
"loss": 2.037,
"step": 23
},
{
"epoch": 0.15874328234807772,
"grad_norm": 7.658245086669922,
"learning_rate": 8.891708613973126e-05,
"loss": 2.3119,
"step": 24
},
{
"epoch": 0.1653575857792476,
"grad_norm": 6.482032775878906,
"learning_rate": 8.788761839251559e-05,
"loss": 2.2091,
"step": 25
},
{
"epoch": 0.1653575857792476,
"eval_loss": 2.0168991088867188,
"eval_runtime": 4.0539,
"eval_samples_per_second": 12.334,
"eval_steps_per_second": 3.207,
"step": 25
},
{
"epoch": 0.17197188921041753,
"grad_norm": 8.424822807312012,
"learning_rate": 8.681980515339464e-05,
"loss": 2.0132,
"step": 26
},
{
"epoch": 0.17858619264158743,
"grad_norm": 3.7590909004211426,
"learning_rate": 8.571489144483944e-05,
"loss": 1.7841,
"step": 27
},
{
"epoch": 0.18520049607275735,
"grad_norm": 1.9769529104232788,
"learning_rate": 8.457416554680877e-05,
"loss": 1.7366,
"step": 28
},
{
"epoch": 0.19181479950392724,
"grad_norm": 1.1488988399505615,
"learning_rate": 8.339895749467238e-05,
"loss": 1.7041,
"step": 29
},
{
"epoch": 0.19842910293509713,
"grad_norm": 3.0149729251861572,
"learning_rate": 8.219063752844926e-05,
"loss": 1.7511,
"step": 30
},
{
"epoch": 0.20504340636626706,
"grad_norm": 4.112295627593994,
"learning_rate": 8.095061449516903e-05,
"loss": 1.9589,
"step": 31
},
{
"epoch": 0.21165770979743695,
"grad_norm": 1.706176519393921,
"learning_rate": 7.968033420621935e-05,
"loss": 1.9123,
"step": 32
},
{
"epoch": 0.21827201322860687,
"grad_norm": 1.1065466403961182,
"learning_rate": 7.838127775159452e-05,
"loss": 1.8454,
"step": 33
},
{
"epoch": 0.22488631665977676,
"grad_norm": 1.1772329807281494,
"learning_rate": 7.705495977301078e-05,
"loss": 1.7768,
"step": 34
},
{
"epoch": 0.23150062009094668,
"grad_norm": 1.597283124923706,
"learning_rate": 7.570292669790186e-05,
"loss": 1.7835,
"step": 35
},
{
"epoch": 0.23811492352211658,
"grad_norm": 2.4821412563323975,
"learning_rate": 7.43267549363537e-05,
"loss": 1.944,
"step": 36
},
{
"epoch": 0.24472922695328647,
"grad_norm": 3.8077642917633057,
"learning_rate": 7.292804904308087e-05,
"loss": 2.0487,
"step": 37
},
{
"epoch": 0.2513435303844564,
"grad_norm": 3.4441919326782227,
"learning_rate": 7.150843984658754e-05,
"loss": 1.972,
"step": 38
},
{
"epoch": 0.2579578338156263,
"grad_norm": 1.2412813901901245,
"learning_rate": 7.006958254769438e-05,
"loss": 1.6434,
"step": 39
},
{
"epoch": 0.2645721372467962,
"grad_norm": 1.4234942197799683,
"learning_rate": 6.861315478964841e-05,
"loss": 1.6448,
"step": 40
},
{
"epoch": 0.2711864406779661,
"grad_norm": 1.5209438800811768,
"learning_rate": 6.714085470206609e-05,
"loss": 1.656,
"step": 41
},
{
"epoch": 0.277800744109136,
"grad_norm": 1.6743462085723877,
"learning_rate": 6.56543989209901e-05,
"loss": 1.6416,
"step": 42
},
{
"epoch": 0.2844150475403059,
"grad_norm": 1.5066150426864624,
"learning_rate": 6.415552058736854e-05,
"loss": 1.7058,
"step": 43
},
{
"epoch": 0.2910293509714758,
"grad_norm": 1.3054348230361938,
"learning_rate": 6.264596732629e-05,
"loss": 1.7489,
"step": 44
},
{
"epoch": 0.29764365440264573,
"grad_norm": 1.1575602293014526,
"learning_rate": 6.112749920933111e-05,
"loss": 1.8122,
"step": 45
},
{
"epoch": 0.30425795783381565,
"grad_norm": 0.9143672585487366,
"learning_rate": 5.960188670239154e-05,
"loss": 1.726,
"step": 46
},
{
"epoch": 0.3108722612649855,
"grad_norm": 1.3133972883224487,
"learning_rate": 5.80709086014102e-05,
"loss": 1.7691,
"step": 47
},
{
"epoch": 0.31748656469615544,
"grad_norm": 1.7563693523406982,
"learning_rate": 5.653634995836856e-05,
"loss": 1.862,
"step": 48
},
{
"epoch": 0.32410086812732536,
"grad_norm": 3.2339706420898438,
"learning_rate": 5.500000000000001e-05,
"loss": 1.9687,
"step": 49
},
{
"epoch": 0.3307151715584952,
"grad_norm": 1.415595531463623,
"learning_rate": 5.346365004163145e-05,
"loss": 1.9923,
"step": 50
},
{
"epoch": 0.3307151715584952,
"eval_loss": 1.8034321069717407,
"eval_runtime": 4.0537,
"eval_samples_per_second": 12.334,
"eval_steps_per_second": 3.207,
"step": 50
},
{
"epoch": 0.33732947498966515,
"grad_norm": 1.7704148292541504,
"learning_rate": 5.192909139858981e-05,
"loss": 1.7264,
"step": 51
},
{
"epoch": 0.34394377842083507,
"grad_norm": 0.8146671056747437,
"learning_rate": 5.0398113297608465e-05,
"loss": 1.593,
"step": 52
},
{
"epoch": 0.350558081852005,
"grad_norm": 0.8015440702438354,
"learning_rate": 4.887250079066892e-05,
"loss": 1.6073,
"step": 53
},
{
"epoch": 0.35717238528317485,
"grad_norm": 0.6833037734031677,
"learning_rate": 4.7354032673710005e-05,
"loss": 1.6055,
"step": 54
},
{
"epoch": 0.3637866887143448,
"grad_norm": 1.2183301448822021,
"learning_rate": 4.584447941263149e-05,
"loss": 1.6739,
"step": 55
},
{
"epoch": 0.3704009921455147,
"grad_norm": 2.2584455013275146,
"learning_rate": 4.43456010790099e-05,
"loss": 1.7502,
"step": 56
},
{
"epoch": 0.37701529557668456,
"grad_norm": 2.1715304851531982,
"learning_rate": 4.285914529793391e-05,
"loss": 1.7755,
"step": 57
},
{
"epoch": 0.3836295990078545,
"grad_norm": 2.027273654937744,
"learning_rate": 4.13868452103516e-05,
"loss": 1.732,
"step": 58
},
{
"epoch": 0.3902439024390244,
"grad_norm": 1.1867992877960205,
"learning_rate": 3.9930417452305626e-05,
"loss": 1.6905,
"step": 59
},
{
"epoch": 0.39685820587019427,
"grad_norm": 1.161856770515442,
"learning_rate": 3.8491560153412466e-05,
"loss": 1.7519,
"step": 60
},
{
"epoch": 0.4034725093013642,
"grad_norm": 2.370361804962158,
"learning_rate": 3.707195095691913e-05,
"loss": 1.8929,
"step": 61
},
{
"epoch": 0.4100868127325341,
"grad_norm": 1.8539221286773682,
"learning_rate": 3.567324506364632e-05,
"loss": 1.8778,
"step": 62
},
{
"epoch": 0.41670111616370403,
"grad_norm": 1.790212631225586,
"learning_rate": 3.4297073302098156e-05,
"loss": 1.7908,
"step": 63
},
{
"epoch": 0.4233154195948739,
"grad_norm": 1.4143702983856201,
"learning_rate": 3.2945040226989244e-05,
"loss": 1.5811,
"step": 64
},
{
"epoch": 0.4299297230260438,
"grad_norm": 1.0017449855804443,
"learning_rate": 3.16187222484055e-05,
"loss": 1.56,
"step": 65
},
{
"epoch": 0.43654402645721374,
"grad_norm": 0.7748380303382874,
"learning_rate": 3.0319665793780648e-05,
"loss": 1.5291,
"step": 66
},
{
"epoch": 0.4431583298883836,
"grad_norm": 0.7852901220321655,
"learning_rate": 2.9049385504830985e-05,
"loss": 1.6028,
"step": 67
},
{
"epoch": 0.4497726333195535,
"grad_norm": 1.797322392463684,
"learning_rate": 2.7809362471550748e-05,
"loss": 1.746,
"step": 68
},
{
"epoch": 0.45638693675072345,
"grad_norm": 2.199897050857544,
"learning_rate": 2.660104250532764e-05,
"loss": 1.7829,
"step": 69
},
{
"epoch": 0.46300124018189337,
"grad_norm": 2.4895544052124023,
"learning_rate": 2.5425834453191232e-05,
"loss": 1.8072,
"step": 70
},
{
"epoch": 0.46961554361306324,
"grad_norm": 1.3517361879348755,
"learning_rate": 2.4285108555160577e-05,
"loss": 1.711,
"step": 71
},
{
"epoch": 0.47622984704423316,
"grad_norm": 1.6711078882217407,
"learning_rate": 2.3180194846605367e-05,
"loss": 1.7425,
"step": 72
},
{
"epoch": 0.4828441504754031,
"grad_norm": 1.413799524307251,
"learning_rate": 2.2112381607484417e-05,
"loss": 1.7813,
"step": 73
},
{
"epoch": 0.48945845390657294,
"grad_norm": 1.5747374296188354,
"learning_rate": 2.1082913860268765e-05,
"loss": 1.8938,
"step": 74
},
{
"epoch": 0.49607275733774286,
"grad_norm": 1.6026586294174194,
"learning_rate": 2.0092991918301108e-05,
"loss": 1.9236,
"step": 75
},
{
"epoch": 0.49607275733774286,
"eval_loss": 1.7268242835998535,
"eval_runtime": 4.0519,
"eval_samples_per_second": 12.34,
"eval_steps_per_second": 3.208,
"step": 75
},
{
"epoch": 0.5026870607689128,
"grad_norm": 1.0837961435317993,
"learning_rate": 1.91437699862843e-05,
"loss": 1.5649,
"step": 76
},
{
"epoch": 0.5093013642000827,
"grad_norm": 1.3891311883926392,
"learning_rate": 1.8236354814530112e-05,
"loss": 1.5706,
"step": 77
},
{
"epoch": 0.5159156676312526,
"grad_norm": 0.9055605530738831,
"learning_rate": 1.7371804408538024e-05,
"loss": 1.5843,
"step": 78
},
{
"epoch": 0.5225299710624225,
"grad_norm": 1.7795766592025757,
"learning_rate": 1.6551126795408016e-05,
"loss": 1.5982,
"step": 79
},
{
"epoch": 0.5291442744935924,
"grad_norm": 1.0256783962249756,
"learning_rate": 1.577527884852619e-05,
"loss": 1.6283,
"step": 80
},
{
"epoch": 0.5357585779247623,
"grad_norm": 1.4087269306182861,
"learning_rate": 1.5045165171893116e-05,
"loss": 1.7221,
"step": 81
},
{
"epoch": 0.5423728813559322,
"grad_norm": 1.3582618236541748,
"learning_rate": 1.4361637045396029e-05,
"loss": 1.7319,
"step": 82
},
{
"epoch": 0.5489871847871021,
"grad_norm": 1.3992999792099,
"learning_rate": 1.3725491432254624e-05,
"loss": 1.7597,
"step": 83
},
{
"epoch": 0.555601488218272,
"grad_norm": 1.0967944860458374,
"learning_rate": 1.313747004979751e-05,
"loss": 1.636,
"step": 84
},
{
"epoch": 0.5622157916494419,
"grad_norm": 1.2128665447235107,
"learning_rate": 1.2598258504653081e-05,
"loss": 1.6875,
"step": 85
},
{
"epoch": 0.5688300950806118,
"grad_norm": 1.3232948780059814,
"learning_rate": 1.2108485493362765e-05,
"loss": 1.8021,
"step": 86
},
{
"epoch": 0.5754443985117818,
"grad_norm": 1.3909993171691895,
"learning_rate": 1.1668722069349041e-05,
"loss": 1.8413,
"step": 87
},
{
"epoch": 0.5820587019429516,
"grad_norm": 1.2057301998138428,
"learning_rate": 1.1279480977092635e-05,
"loss": 1.73,
"step": 88
},
{
"epoch": 0.5886730053741215,
"grad_norm": 0.6822798252105713,
"learning_rate": 1.094121605429547e-05,
"loss": 1.5085,
"step": 89
},
{
"epoch": 0.5952873088052915,
"grad_norm": 0.7973278760910034,
"learning_rate": 1.0654321702726141e-05,
"loss": 1.5389,
"step": 90
},
{
"epoch": 0.6019016122364613,
"grad_norm": 0.8194143772125244,
"learning_rate": 1.0419132428365116e-05,
"loss": 1.563,
"step": 91
},
{
"epoch": 0.6085159156676313,
"grad_norm": 0.8316376209259033,
"learning_rate": 1.0235922451385733e-05,
"loss": 1.5623,
"step": 92
},
{
"epoch": 0.6151302190988012,
"grad_norm": 1.0160751342773438,
"learning_rate": 1.0104905386425733e-05,
"loss": 1.6195,
"step": 93
},
{
"epoch": 0.621744522529971,
"grad_norm": 1.2698185443878174,
"learning_rate": 1.002623399352217e-05,
"loss": 1.6597,
"step": 94
},
{
"epoch": 0.628358825961141,
"grad_norm": 0.7779133319854736,
"learning_rate": 1e-05,
"loss": 1.6596,
"step": 95
}
],
"logging_steps": 1,
"max_steps": 95,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 25,
"stateful_callbacks": {
"EarlyStoppingCallback": {
"args": {
"early_stopping_patience": 1,
"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": 1.1698376402502943e+18,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}