{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 500,
"global_step": 76,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.013157894736842105,
"grad_norm": 1699.4203999374324,
"learning_rate": 1.25e-06,
"loss": 11.6905,
"step": 1
},
{
"epoch": 0.02631578947368421,
"grad_norm": 1649.6649286698323,
"learning_rate": 2.5e-06,
"loss": 11.7067,
"step": 2
},
{
"epoch": 0.039473684210526314,
"grad_norm": 1109.6668329617996,
"learning_rate": 3.7500000000000005e-06,
"loss": 10.8238,
"step": 3
},
{
"epoch": 0.05263157894736842,
"grad_norm": 1106.178486585862,
"learning_rate": 5e-06,
"loss": 9.0694,
"step": 4
},
{
"epoch": 0.06578947368421052,
"grad_norm": 809.3294910843833,
"learning_rate": 6.25e-06,
"loss": 5.9203,
"step": 5
},
{
"epoch": 0.07894736842105263,
"grad_norm": 257.0422697393994,
"learning_rate": 7.500000000000001e-06,
"loss": 4.7981,
"step": 6
},
{
"epoch": 0.09210526315789473,
"grad_norm": 394.7803420902946,
"learning_rate": 8.750000000000001e-06,
"loss": 5.5132,
"step": 7
},
{
"epoch": 0.10526315789473684,
"grad_norm": 126.60762803235828,
"learning_rate": 1e-05,
"loss": 3.2079,
"step": 8
},
{
"epoch": 0.11842105263157894,
"grad_norm": 125.30453375076969,
"learning_rate": 9.994664874011864e-06,
"loss": 5.0667,
"step": 9
},
{
"epoch": 0.13157894736842105,
"grad_norm": 158.38384375765978,
"learning_rate": 9.978670881475173e-06,
"loss": 4.1611,
"step": 10
},
{
"epoch": 0.14473684210526316,
"grad_norm": 55.659050974987274,
"learning_rate": 9.952052154376027e-06,
"loss": 3.4048,
"step": 11
},
{
"epoch": 0.15789473684210525,
"grad_norm": 42.873598431489626,
"learning_rate": 9.91486549841951e-06,
"loss": 4.0195,
"step": 12
},
{
"epoch": 0.17105263157894737,
"grad_norm": 72.14863137487674,
"learning_rate": 9.867190271803466e-06,
"loss": 4.0778,
"step": 13
},
{
"epoch": 0.18421052631578946,
"grad_norm": 39.55596624092672,
"learning_rate": 9.809128215864096e-06,
"loss": 3.8024,
"step": 14
},
{
"epoch": 0.19736842105263158,
"grad_norm": 18.04964705613809,
"learning_rate": 9.74080323795483e-06,
"loss": 3.3671,
"step": 15
},
{
"epoch": 0.21052631578947367,
"grad_norm": 14.28948406963546,
"learning_rate": 9.66236114702178e-06,
"loss": 2.7161,
"step": 16
},
{
"epoch": 0.2236842105263158,
"grad_norm": 14.98902092554103,
"learning_rate": 9.573969342440107e-06,
"loss": 2.6427,
"step": 17
},
{
"epoch": 0.23684210526315788,
"grad_norm": 17.857073526805326,
"learning_rate": 9.475816456775313e-06,
"loss": 2.5664,
"step": 18
},
{
"epoch": 0.25,
"grad_norm": 17.982906576778536,
"learning_rate": 9.368111953231849e-06,
"loss": 2.3566,
"step": 19
},
{
"epoch": 0.2631578947368421,
"grad_norm": 14.066497831682323,
"learning_rate": 9.251085678648072e-06,
"loss": 2.3749,
"step": 20
},
{
"epoch": 0.27631578947368424,
"grad_norm": 16.568001275060062,
"learning_rate": 9.124987372991512e-06,
"loss": 2.3824,
"step": 21
},
{
"epoch": 0.2894736842105263,
"grad_norm": 10.225207891248253,
"learning_rate": 8.990086136401199e-06,
"loss": 2.2876,
"step": 22
},
{
"epoch": 0.3026315789473684,
"grad_norm": 15.52475147772328,
"learning_rate": 8.846669854914395e-06,
"loss": 2.0579,
"step": 23
},
{
"epoch": 0.3157894736842105,
"grad_norm": 10.735876859280078,
"learning_rate": 8.695044586103297e-06,
"loss": 2.1149,
"step": 24
},
{
"epoch": 0.32894736842105265,
"grad_norm": 12.03589457222836,
"learning_rate": 8.535533905932739e-06,
"loss": 2.0015,
"step": 25
},
{
"epoch": 0.34210526315789475,
"grad_norm": 10.668355704543856,
"learning_rate": 8.368478218232787e-06,
"loss": 2.0109,
"step": 26
},
{
"epoch": 0.35526315789473684,
"grad_norm": 11.247848100294151,
"learning_rate": 8.194234028259806e-06,
"loss": 2.0085,
"step": 27
},
{
"epoch": 0.3684210526315789,
"grad_norm": 10.041875020749039,
"learning_rate": 8.013173181896283e-06,
"loss": 1.707,
"step": 28
},
{
"epoch": 0.3815789473684211,
"grad_norm": 16.66505157922524,
"learning_rate": 7.82568207211296e-06,
"loss": 1.9144,
"step": 29
},
{
"epoch": 0.39473684210526316,
"grad_norm": 14.129248059213934,
"learning_rate": 7.63216081438678e-06,
"loss": 1.7373,
"step": 30
},
{
"epoch": 0.40789473684210525,
"grad_norm": 8.400365715854278,
"learning_rate": 7.4330223928342814e-06,
"loss": 1.606,
"step": 31
},
{
"epoch": 0.42105263157894735,
"grad_norm": 9.454647164990588,
"learning_rate": 7.2286917788826926e-06,
"loss": 1.5251,
"step": 32
},
{
"epoch": 0.4342105263157895,
"grad_norm": 11.708573459019293,
"learning_rate": 7.019605024359475e-06,
"loss": 1.4462,
"step": 33
},
{
"epoch": 0.4473684210526316,
"grad_norm": 11.288211975525503,
"learning_rate": 6.806208330935766e-06,
"loss": 1.3242,
"step": 34
},
{
"epoch": 0.4605263157894737,
"grad_norm": 12.84833845270528,
"learning_rate": 6.588957097909509e-06,
"loss": 1.2777,
"step": 35
},
{
"epoch": 0.47368421052631576,
"grad_norm": 12.805717587983919,
"learning_rate": 6.368314950360416e-06,
"loss": 1.2439,
"step": 36
},
{
"epoch": 0.4868421052631579,
"grad_norm": 16.79814160674569,
"learning_rate": 6.144752749750671e-06,
"loss": 1.2062,
"step": 37
},
{
"epoch": 0.5,
"grad_norm": 11.704864343239466,
"learning_rate": 5.918747589082853e-06,
"loss": 1.2953,
"step": 38
},
{
"epoch": 0.5131578947368421,
"grad_norm": 9.088068501685061,
"learning_rate": 5.690781774759412e-06,
"loss": 1.1413,
"step": 39
},
{
"epoch": 0.5263157894736842,
"grad_norm": 9.452972340503731,
"learning_rate": 5.46134179731651e-06,
"loss": 1.1089,
"step": 40
},
{
"epoch": 0.5394736842105263,
"grad_norm": 13.866484586191572,
"learning_rate": 5.230917293228699e-06,
"loss": 1.0386,
"step": 41
},
{
"epoch": 0.5526315789473685,
"grad_norm": 9.961380544449332,
"learning_rate": 5e-06,
"loss": 1.0679,
"step": 42
},
{
"epoch": 0.5657894736842105,
"grad_norm": 8.460862397122703,
"learning_rate": 4.7690827067713035e-06,
"loss": 1.1567,
"step": 43
},
{
"epoch": 0.5789473684210527,
"grad_norm": 8.823619822316715,
"learning_rate": 4.53865820268349e-06,
"loss": 1.049,
"step": 44
},
{
"epoch": 0.5921052631578947,
"grad_norm": 7.271462434248026,
"learning_rate": 4.309218225240591e-06,
"loss": 0.9448,
"step": 45
},
{
"epoch": 0.6052631578947368,
"grad_norm": 15.732461239193407,
"learning_rate": 4.081252410917148e-06,
"loss": 0.9914,
"step": 46
},
{
"epoch": 0.618421052631579,
"grad_norm": 11.654049140402229,
"learning_rate": 3.855247250249331e-06,
"loss": 0.8027,
"step": 47
},
{
"epoch": 0.631578947368421,
"grad_norm": 17.59370496573567,
"learning_rate": 3.6316850496395863e-06,
"loss": 1.0106,
"step": 48
},
{
"epoch": 0.6447368421052632,
"grad_norm": 10.035162483746694,
"learning_rate": 3.4110429020904924e-06,
"loss": 0.8962,
"step": 49
},
{
"epoch": 0.6578947368421053,
"grad_norm": 7.517763142577863,
"learning_rate": 3.1937916690642356e-06,
"loss": 0.9492,
"step": 50
},
{
"epoch": 0.6710526315789473,
"grad_norm": 15.969988377607551,
"learning_rate": 2.980394975640526e-06,
"loss": 0.8697,
"step": 51
},
{
"epoch": 0.6842105263157895,
"grad_norm": 7.9971953326811125,
"learning_rate": 2.771308221117309e-06,
"loss": 0.9477,
"step": 52
},
{
"epoch": 0.6973684210526315,
"grad_norm": 9.166442366719266,
"learning_rate": 2.5669776071657194e-06,
"loss": 0.9585,
"step": 53
},
{
"epoch": 0.7105263157894737,
"grad_norm": 8.283342586733458,
"learning_rate": 2.3678391856132203e-06,
"loss": 0.948,
"step": 54
},
{
"epoch": 0.7236842105263158,
"grad_norm": 9.135125773249728,
"learning_rate": 2.174317927887041e-06,
"loss": 0.8435,
"step": 55
},
{
"epoch": 0.7368421052631579,
"grad_norm": 7.738718289232403,
"learning_rate": 1.9868268181037186e-06,
"loss": 0.9933,
"step": 56
},
{
"epoch": 0.75,
"grad_norm": 5.632776555168388,
"learning_rate": 1.8057659717401948e-06,
"loss": 0.8011,
"step": 57
},
{
"epoch": 0.7631578947368421,
"grad_norm": 6.642349176520887,
"learning_rate": 1.6315217817672142e-06,
"loss": 0.7914,
"step": 58
},
{
"epoch": 0.7763157894736842,
"grad_norm": 6.204837918911468,
"learning_rate": 1.4644660940672628e-06,
"loss": 0.7612,
"step": 59
},
{
"epoch": 0.7894736842105263,
"grad_norm": 6.881328834033968,
"learning_rate": 1.3049554138967052e-06,
"loss": 0.7503,
"step": 60
},
{
"epoch": 0.8026315789473685,
"grad_norm": 6.131564640099481,
"learning_rate": 1.1533301450856054e-06,
"loss": 0.817,
"step": 61
},
{
"epoch": 0.8157894736842105,
"grad_norm": 5.8740721419365585,
"learning_rate": 1.0099138635988026e-06,
"loss": 0.7754,
"step": 62
},
{
"epoch": 0.8289473684210527,
"grad_norm": 5.434760658098561,
"learning_rate": 8.750126270084891e-07,
"loss": 0.8402,
"step": 63
},
{
"epoch": 0.8421052631578947,
"grad_norm": 5.408440691228196,
"learning_rate": 7.489143213519301e-07,
"loss": 0.7445,
"step": 64
},
{
"epoch": 0.8552631578947368,
"grad_norm": 4.506252846496264,
"learning_rate": 6.318880467681527e-07,
"loss": 0.6984,
"step": 65
},
{
"epoch": 0.868421052631579,
"grad_norm": 7.264506467869279,
"learning_rate": 5.241835432246888e-07,
"loss": 0.7549,
"step": 66
},
{
"epoch": 0.881578947368421,
"grad_norm": 5.671765225216593,
"learning_rate": 4.2603065755989493e-07,
"loss": 0.8249,
"step": 67
},
{
"epoch": 0.8947368421052632,
"grad_norm": 4.946461980264793,
"learning_rate": 3.3763885297822153e-07,
"loss": 0.7372,
"step": 68
},
{
"epoch": 0.9078947368421053,
"grad_norm": 7.216022490434719,
"learning_rate": 2.5919676204517073e-07,
"loss": 0.7816,
"step": 69
},
{
"epoch": 0.9210526315789473,
"grad_norm": 4.627827587948695,
"learning_rate": 1.908717841359048e-07,
"loss": 0.786,
"step": 70
},
{
"epoch": 0.9342105263157895,
"grad_norm": 3.6873839890732363,
"learning_rate": 1.328097281965357e-07,
"loss": 0.7109,
"step": 71
},
{
"epoch": 0.9473684210526315,
"grad_norm": 3.9116965099593646,
"learning_rate": 8.513450158049109e-08,
"loss": 0.7468,
"step": 72
},
{
"epoch": 0.9605263157894737,
"grad_norm": 4.8726672621846365,
"learning_rate": 4.794784562397459e-08,
"loss": 0.6888,
"step": 73
},
{
"epoch": 0.9736842105263158,
"grad_norm": 4.616871357784515,
"learning_rate": 2.1329118524827662e-08,
"loss": 0.7657,
"step": 74
},
{
"epoch": 0.9868421052631579,
"grad_norm": 4.101276371849697,
"learning_rate": 5.3351259881379016e-09,
"loss": 0.686,
"step": 75
},
{
"epoch": 1.0,
"grad_norm": 4.36368396992744,
"learning_rate": 0.0,
"loss": 0.8604,
"step": 76
},
{
"epoch": 1.0,
"eval_loss": 0.7618905305862427,
"eval_runtime": 129.5439,
"eval_samples_per_second": 39.407,
"eval_steps_per_second": 1.235,
"step": 76
},
{
"epoch": 1.0,
"step": 76,
"total_flos": 2.244031216091136e+16,
"train_loss": 2.2036053005017733,
"train_runtime": 949.5806,
"train_samples_per_second": 10.213,
"train_steps_per_second": 0.08
}
],
"logging_steps": 1,
"max_steps": 76,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"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": 2.244031216091136e+16,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}