sanity_syntax / trainer_state.json
terry69's picture
Model save
ee16748 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9957567185289957,
"eval_steps": 500,
"global_step": 176,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.005657708628005658,
"grad_norm": 0.43130557078322906,
"learning_rate": 1.1111111111111112e-05,
"loss": 1.9358,
"step": 1
},
{
"epoch": 0.028288543140028287,
"grad_norm": 0.47181074868090295,
"learning_rate": 5.555555555555556e-05,
"loss": 1.8755,
"step": 5
},
{
"epoch": 0.056577086280056574,
"grad_norm": 0.24499816695787435,
"learning_rate": 0.00011111111111111112,
"loss": 1.8709,
"step": 10
},
{
"epoch": 0.08486562942008487,
"grad_norm": 0.1946906656901304,
"learning_rate": 0.0001666666666666667,
"loss": 1.783,
"step": 15
},
{
"epoch": 0.11315417256011315,
"grad_norm": 0.16173800943880282,
"learning_rate": 0.00019992093972273018,
"loss": 1.6992,
"step": 20
},
{
"epoch": 0.14144271570014144,
"grad_norm": 0.13810234079187747,
"learning_rate": 0.0001990329466472502,
"loss": 1.6291,
"step": 25
},
{
"epoch": 0.16973125884016974,
"grad_norm": 0.11800201667039435,
"learning_rate": 0.000197166934004041,
"loss": 1.6379,
"step": 30
},
{
"epoch": 0.19801980198019803,
"grad_norm": 0.11256043626490136,
"learning_rate": 0.00019434132997221345,
"loss": 1.5736,
"step": 35
},
{
"epoch": 0.2263083451202263,
"grad_norm": 0.10718182461540707,
"learning_rate": 0.00019058403936655233,
"loss": 1.5491,
"step": 40
},
{
"epoch": 0.2545968882602546,
"grad_norm": 0.112326110136943,
"learning_rate": 0.00018593216805796612,
"loss": 1.51,
"step": 45
},
{
"epoch": 0.2828854314002829,
"grad_norm": 0.08515607635399414,
"learning_rate": 0.00018043165652707649,
"loss": 1.5018,
"step": 50
},
{
"epoch": 0.31117397454031115,
"grad_norm": 0.08598762497133733,
"learning_rate": 0.00017413682616986185,
"loss": 1.4985,
"step": 55
},
{
"epoch": 0.33946251768033947,
"grad_norm": 0.08605574729719911,
"learning_rate": 0.0001671098428359037,
"loss": 1.4957,
"step": 60
},
{
"epoch": 0.36775106082036774,
"grad_norm": 0.08270099201021953,
"learning_rate": 0.00015942010289717105,
"loss": 1.4897,
"step": 65
},
{
"epoch": 0.39603960396039606,
"grad_norm": 0.09139373323353543,
"learning_rate": 0.00015114354791034225,
"loss": 1.4661,
"step": 70
},
{
"epoch": 0.4243281471004243,
"grad_norm": 0.09213917858670279,
"learning_rate": 0.00014236191464085286,
"loss": 1.4825,
"step": 75
},
{
"epoch": 0.4526166902404526,
"grad_norm": 0.09741017460001462,
"learning_rate": 0.0001331619278552068,
"loss": 1.4796,
"step": 80
},
{
"epoch": 0.4809052333804809,
"grad_norm": 0.09522302084100907,
"learning_rate": 0.0001236344438532905,
"loss": 1.4522,
"step": 85
},
{
"epoch": 0.5091937765205092,
"grad_norm": 0.10005523175818275,
"learning_rate": 0.00011387355319890685,
"loss": 1.4512,
"step": 90
},
{
"epoch": 0.5374823196605375,
"grad_norm": 0.09774894920042008,
"learning_rate": 0.0001039756515096926,
"loss": 1.4531,
"step": 95
},
{
"epoch": 0.5657708628005658,
"grad_norm": 0.09337090709938024,
"learning_rate": 9.403848748301802e-05,
"loss": 1.431,
"step": 100
},
{
"epoch": 0.594059405940594,
"grad_norm": 0.10862536785181011,
"learning_rate": 8.416019755927851e-05,
"loss": 1.4497,
"step": 105
},
{
"epoch": 0.6223479490806223,
"grad_norm": 0.09931981935895794,
"learning_rate": 7.443833675595255e-05,
"loss": 1.4277,
"step": 110
},
{
"epoch": 0.6506364922206507,
"grad_norm": 0.09784818782800696,
"learning_rate": 6.496891524361757e-05,
"loss": 1.4467,
"step": 115
},
{
"epoch": 0.6789250353606789,
"grad_norm": 0.09291305679558234,
"learning_rate": 5.584545017840885e-05,
"loss": 1.4384,
"step": 120
},
{
"epoch": 0.7072135785007072,
"grad_norm": 0.10423529826360275,
"learning_rate": 4.715804215473809e-05,
"loss": 1.4476,
"step": 125
},
{
"epoch": 0.7355021216407355,
"grad_norm": 0.10820562792911577,
"learning_rate": 3.899248539894757e-05,
"loss": 1.4327,
"step": 130
},
{
"epoch": 0.7637906647807637,
"grad_norm": 0.09010400983561788,
"learning_rate": 3.14294204913587e-05,
"loss": 1.4426,
"step": 135
},
{
"epoch": 0.7920792079207921,
"grad_norm": 0.093194757033809,
"learning_rate": 2.4543537984176978e-05,
"loss": 1.4434,
"step": 140
},
{
"epoch": 0.8203677510608204,
"grad_norm": 0.09797367971916428,
"learning_rate": 1.840284078008393e-05,
"loss": 1.4363,
"step": 145
},
{
"epoch": 0.8486562942008486,
"grad_norm": 0.09241502492056575,
"learning_rate": 1.3067972556041752e-05,
"loss": 1.4347,
"step": 150
},
{
"epoch": 0.8769448373408769,
"grad_norm": 0.0929511573879462,
"learning_rate": 8.59161886459654e-06,
"loss": 1.4129,
"step": 155
},
{
"epoch": 0.9052333804809052,
"grad_norm": 0.08984575236333624,
"learning_rate": 5.017986827221733e-06,
"loss": 1.4555,
"step": 160
},
{
"epoch": 0.9335219236209336,
"grad_norm": 0.0905814734943296,
"learning_rate": 2.3823685580949273e-06,
"loss": 1.4418,
"step": 165
},
{
"epoch": 0.9618104667609618,
"grad_norm": 0.08926337770672606,
"learning_rate": 7.10792629802659e-07,
"loss": 1.4399,
"step": 170
},
{
"epoch": 0.9900990099009901,
"grad_norm": 0.09160289967908435,
"learning_rate": 1.976702299344435e-08,
"loss": 1.4333,
"step": 175
},
{
"epoch": 0.9957567185289957,
"eval_loss": 1.235967755317688,
"eval_runtime": 1555.7732,
"eval_samples_per_second": 8.599,
"eval_steps_per_second": 0.538,
"step": 176
},
{
"epoch": 0.9957567185289957,
"step": 176,
"total_flos": 2207623390691328.0,
"train_loss": 1.0380965369668873,
"train_runtime": 4181.7688,
"train_samples_per_second": 2.703,
"train_steps_per_second": 0.042
}
],
"logging_steps": 5,
"max_steps": 176,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 25,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 2207623390691328.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}