ZhangShenao's picture
Model save
24bfe31 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 6.461538461538462,
"eval_steps": 500,
"global_step": 63,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.10256410256410256,
"grad_norm": 45.74676831817011,
"learning_rate": 2.8571428571428573e-06,
"loss": 1.7736,
"step": 1
},
{
"epoch": 0.20512820512820512,
"grad_norm": 43.8997476738742,
"learning_rate": 5.7142857142857145e-06,
"loss": 1.7436,
"step": 2
},
{
"epoch": 0.3076923076923077,
"grad_norm": 33.1407433311641,
"learning_rate": 8.571428571428571e-06,
"loss": 1.5701,
"step": 3
},
{
"epoch": 0.41025641025641024,
"grad_norm": 15.347997497580668,
"learning_rate": 1.1428571428571429e-05,
"loss": 1.3552,
"step": 4
},
{
"epoch": 0.5128205128205128,
"grad_norm": 42.292432586275865,
"learning_rate": 1.4285714285714287e-05,
"loss": 1.2627,
"step": 5
},
{
"epoch": 0.6153846153846154,
"grad_norm": 36.376951471898245,
"learning_rate": 1.7142857142857142e-05,
"loss": 1.2198,
"step": 6
},
{
"epoch": 0.717948717948718,
"grad_norm": 7.974991090529373,
"learning_rate": 2e-05,
"loss": 1.1377,
"step": 7
},
{
"epoch": 0.8205128205128205,
"grad_norm": 16.501115024701512,
"learning_rate": 1.998426815017817e-05,
"loss": 1.1813,
"step": 8
},
{
"epoch": 0.9230769230769231,
"grad_norm": 7.188262087017176,
"learning_rate": 1.9937122098932428e-05,
"loss": 1.1565,
"step": 9
},
{
"epoch": 1.0256410256410255,
"grad_norm": 4.140303367679906,
"learning_rate": 1.985871018518236e-05,
"loss": 1.03,
"step": 10
},
{
"epoch": 1.1282051282051282,
"grad_norm": 2.728109728903398,
"learning_rate": 1.9749279121818235e-05,
"loss": 0.9516,
"step": 11
},
{
"epoch": 1.2307692307692308,
"grad_norm": 5.943582967422805,
"learning_rate": 1.9609173219450998e-05,
"loss": 0.9511,
"step": 12
},
{
"epoch": 1.3333333333333333,
"grad_norm": 2.215718821444342,
"learning_rate": 1.9438833303083677e-05,
"loss": 0.927,
"step": 13
},
{
"epoch": 1.435897435897436,
"grad_norm": 3.7476553781843758,
"learning_rate": 1.9238795325112867e-05,
"loss": 0.9357,
"step": 14
},
{
"epoch": 1.5384615384615383,
"grad_norm": 2.3784620064293667,
"learning_rate": 1.900968867902419e-05,
"loss": 0.9271,
"step": 15
},
{
"epoch": 1.641025641025641,
"grad_norm": 1.921424026693961,
"learning_rate": 1.8752234219087538e-05,
"loss": 0.9106,
"step": 16
},
{
"epoch": 1.7435897435897436,
"grad_norm": 1.74875556144991,
"learning_rate": 1.8467241992282842e-05,
"loss": 0.8924,
"step": 17
},
{
"epoch": 1.8461538461538463,
"grad_norm": 1.5465709392794136,
"learning_rate": 1.8155608689592604e-05,
"loss": 0.8985,
"step": 18
},
{
"epoch": 1.9487179487179487,
"grad_norm": 1.6088170768182135,
"learning_rate": 1.78183148246803e-05,
"loss": 0.8731,
"step": 19
},
{
"epoch": 2.051282051282051,
"grad_norm": 1.550794028416961,
"learning_rate": 1.7456421648831658e-05,
"loss": 0.8108,
"step": 20
},
{
"epoch": 2.1538461538461537,
"grad_norm": 1.1891604452468383,
"learning_rate": 1.7071067811865477e-05,
"loss": 0.7693,
"step": 21
},
{
"epoch": 2.2564102564102564,
"grad_norm": 1.7874206012739782,
"learning_rate": 1.6663465779520042e-05,
"loss": 0.7545,
"step": 22
},
{
"epoch": 2.358974358974359,
"grad_norm": 1.546137200179085,
"learning_rate": 1.6234898018587336e-05,
"loss": 0.7505,
"step": 23
},
{
"epoch": 2.4615384615384617,
"grad_norm": 1.5723321757671156,
"learning_rate": 1.578671296179806e-05,
"loss": 0.7592,
"step": 24
},
{
"epoch": 2.564102564102564,
"grad_norm": 1.147889286365701,
"learning_rate": 1.5320320765153367e-05,
"loss": 0.737,
"step": 25
},
{
"epoch": 2.6666666666666665,
"grad_norm": 1.2373533333897078,
"learning_rate": 1.4837188871052399e-05,
"loss": 0.7165,
"step": 26
},
{
"epoch": 2.769230769230769,
"grad_norm": 1.1523919084408827,
"learning_rate": 1.4338837391175582e-05,
"loss": 0.6998,
"step": 27
},
{
"epoch": 2.871794871794872,
"grad_norm": 1.1820135421697542,
"learning_rate": 1.3826834323650899e-05,
"loss": 0.7086,
"step": 28
},
{
"epoch": 2.9743589743589745,
"grad_norm": 1.163468789617511,
"learning_rate": 1.3302790619551673e-05,
"loss": 0.7175,
"step": 29
},
{
"epoch": 3.076923076923077,
"grad_norm": 1.7767072503264743,
"learning_rate": 1.2768355114248493e-05,
"loss": 0.593,
"step": 30
},
{
"epoch": 3.1794871794871793,
"grad_norm": 1.81233019892104,
"learning_rate": 1.2225209339563144e-05,
"loss": 0.5519,
"step": 31
},
{
"epoch": 3.282051282051282,
"grad_norm": 2.1736744456228694,
"learning_rate": 1.1675062233047365e-05,
"loss": 0.5437,
"step": 32
},
{
"epoch": 3.3846153846153846,
"grad_norm": 2.523447078328654,
"learning_rate": 1.1119644761033079e-05,
"loss": 0.5334,
"step": 33
},
{
"epoch": 3.4871794871794872,
"grad_norm": 1.736445380863662,
"learning_rate": 1.0560704472371919e-05,
"loss": 0.5068,
"step": 34
},
{
"epoch": 3.58974358974359,
"grad_norm": 1.4506284420650815,
"learning_rate": 1e-05,
"loss": 0.5092,
"step": 35
},
{
"epoch": 3.6923076923076925,
"grad_norm": 1.4702124114670763,
"learning_rate": 9.439295527628083e-06,
"loss": 0.4965,
"step": 36
},
{
"epoch": 3.7948717948717947,
"grad_norm": 1.460168348735874,
"learning_rate": 8.880355238966923e-06,
"loss": 0.4902,
"step": 37
},
{
"epoch": 3.8974358974358974,
"grad_norm": 1.2827679925297792,
"learning_rate": 8.324937766952638e-06,
"loss": 0.4923,
"step": 38
},
{
"epoch": 4.0,
"grad_norm": 1.3138838652676037,
"learning_rate": 7.774790660436857e-06,
"loss": 0.4731,
"step": 39
},
{
"epoch": 4.102564102564102,
"grad_norm": 2.82702405107918,
"learning_rate": 7.2316448857515076e-06,
"loss": 0.377,
"step": 40
},
{
"epoch": 4.205128205128205,
"grad_norm": 2.2541347262792897,
"learning_rate": 6.697209380448333e-06,
"loss": 0.3562,
"step": 41
},
{
"epoch": 4.3076923076923075,
"grad_norm": 2.5922750621707293,
"learning_rate": 6.173165676349103e-06,
"loss": 0.3482,
"step": 42
},
{
"epoch": 4.410256410256411,
"grad_norm": 3.552401632591804,
"learning_rate": 5.66116260882442e-06,
"loss": 0.3332,
"step": 43
},
{
"epoch": 4.512820512820513,
"grad_norm": 2.5996491260559615,
"learning_rate": 5.1628111289476025e-06,
"loss": 0.3292,
"step": 44
},
{
"epoch": 4.615384615384615,
"grad_norm": 2.0225922587821805,
"learning_rate": 4.679679234846636e-06,
"loss": 0.3134,
"step": 45
},
{
"epoch": 4.717948717948718,
"grad_norm": 1.794698781237276,
"learning_rate": 4.213287038201943e-06,
"loss": 0.3144,
"step": 46
},
{
"epoch": 4.82051282051282,
"grad_norm": 1.78570038657741,
"learning_rate": 3.7651019814126656e-06,
"loss": 0.3054,
"step": 47
},
{
"epoch": 4.923076923076923,
"grad_norm": 1.855808340157979,
"learning_rate": 3.3365342204799613e-06,
"loss": 0.3,
"step": 48
},
{
"epoch": 5.0256410256410255,
"grad_norm": 1.9551127034348588,
"learning_rate": 2.9289321881345257e-06,
"loss": 0.2895,
"step": 49
},
{
"epoch": 5.128205128205128,
"grad_norm": 1.8579864077346009,
"learning_rate": 2.5435783511683444e-06,
"loss": 0.2339,
"step": 50
},
{
"epoch": 5.230769230769231,
"grad_norm": 1.624477309859244,
"learning_rate": 2.1816851753197023e-06,
"loss": 0.2185,
"step": 51
},
{
"epoch": 5.333333333333333,
"grad_norm": 1.7017286954162238,
"learning_rate": 1.8443913104073984e-06,
"loss": 0.2273,
"step": 52
},
{
"epoch": 5.435897435897436,
"grad_norm": 1.612088436018407,
"learning_rate": 1.5327580077171589e-06,
"loss": 0.2157,
"step": 53
},
{
"epoch": 5.538461538461538,
"grad_norm": 1.6667785081621305,
"learning_rate": 1.2477657809124632e-06,
"loss": 0.2163,
"step": 54
},
{
"epoch": 5.641025641025641,
"grad_norm": 1.5501897496769197,
"learning_rate": 9.903113209758098e-07,
"loss": 0.2134,
"step": 55
},
{
"epoch": 5.743589743589744,
"grad_norm": 1.6228058819139755,
"learning_rate": 7.612046748871327e-07,
"loss": 0.2247,
"step": 56
},
{
"epoch": 5.846153846153846,
"grad_norm": 1.4413136220349998,
"learning_rate": 5.611666969163243e-07,
"loss": 0.2157,
"step": 57
},
{
"epoch": 5.948717948717949,
"grad_norm": 1.3872967872162942,
"learning_rate": 3.908267805490051e-07,
"loss": 0.2111,
"step": 58
},
{
"epoch": 6.051282051282051,
"grad_norm": 1.5157677115577846,
"learning_rate": 2.507208781817638e-07,
"loss": 0.1947,
"step": 59
},
{
"epoch": 6.153846153846154,
"grad_norm": 1.5908969907868264,
"learning_rate": 1.4128981481764115e-07,
"loss": 0.1848,
"step": 60
},
{
"epoch": 6.256410256410256,
"grad_norm": 1.5844350731835435,
"learning_rate": 6.287790106757396e-08,
"loss": 0.1888,
"step": 61
},
{
"epoch": 6.358974358974359,
"grad_norm": 1.6757546348969776,
"learning_rate": 1.5731849821833955e-08,
"loss": 0.1826,
"step": 62
},
{
"epoch": 6.461538461538462,
"grad_norm": 1.5614630318772702,
"learning_rate": 0.0,
"loss": 0.1757,
"step": 63
},
{
"epoch": 6.461538461538462,
"step": 63,
"total_flos": 13164745850880.0,
"train_loss": 0.6425543285551525,
"train_runtime": 766.0702,
"train_samples_per_second": 5.647,
"train_steps_per_second": 0.082
}
],
"logging_steps": 1,
"max_steps": 63,
"num_input_tokens_seen": 0,
"num_train_epochs": 7,
"save_steps": 999999,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": false,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 13164745850880.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}