pythia-12b-draft-70m-code / trainer_state.json
knu-mhan's picture
update
99f3ceb
Raw
History Blame Contribute Delete
4.75 kB
Invalid JSON:Unexpected token 'I', ..."ad_norm": Infinity, "... is not valid JSON
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"eval_steps": 500,
"global_step": 222,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.13513513513513514,
"grad_norm": Infinity,
"learning_rate": 3e-05,
"loss": 1.435,
"step": 10
},
{
"epoch": 0.2702702702702703,
"grad_norm": 47.6257438659668,
"learning_rate": 2.9840150521126656e-05,
"loss": 1.4664,
"step": 20
},
{
"epoch": 0.40540540540540543,
"grad_norm": 28.86147689819336,
"learning_rate": 2.936400899862609e-05,
"loss": 1.3927,
"step": 30
},
{
"epoch": 0.5405405405405406,
"grad_norm": 31.34147071838379,
"learning_rate": 2.8581723562397203e-05,
"loss": 1.3359,
"step": 40
},
{
"epoch": 0.6756756756756757,
"grad_norm": 40.83176803588867,
"learning_rate": 2.7509967268348168e-05,
"loss": 1.2689,
"step": 50
},
{
"epoch": 0.8108108108108109,
"grad_norm": 28.7069034576416,
"learning_rate": 2.617158274115673e-05,
"loss": 1.3124,
"step": 60
},
{
"epoch": 0.9459459459459459,
"grad_norm": 31.556358337402344,
"learning_rate": 2.459509532338337e-05,
"loss": 1.1676,
"step": 70
},
{
"epoch": 1.0810810810810811,
"grad_norm": 30.21744155883789,
"learning_rate": 2.2814105107318955e-05,
"loss": 1.178,
"step": 80
},
{
"epoch": 1.2162162162162162,
"grad_norm": 24.259174346923828,
"learning_rate": 2.086657080735234e-05,
"loss": 1.1852,
"step": 90
},
{
"epoch": 1.3513513513513513,
"grad_norm": 19.32608413696289,
"learning_rate": 1.879400073587521e-05,
"loss": 1.0562,
"step": 100
},
{
"epoch": 1.4864864864864864,
"grad_norm": 28.069013595581055,
"learning_rate": 1.664056812566812e-05,
"loss": 1.0509,
"step": 110
},
{
"epoch": 1.6216216216216215,
"grad_norm": 25.697093963623047,
"learning_rate": 1.4452169654135116e-05,
"loss": 1.1403,
"step": 120
},
{
"epoch": 1.7567567567567568,
"grad_norm": 21.39369773864746,
"learning_rate": 1.2275447235308454e-05,
"loss": 1.1449,
"step": 130
},
{
"epoch": 1.8918918918918919,
"grad_norm": 22.039438247680664,
"learning_rate": 1.015679392842831e-05,
"loss": 1.0383,
"step": 140
},
{
"epoch": 2.027027027027027,
"grad_norm": 25.26503562927246,
"learning_rate": 8.141365150430423e-06,
"loss": 1.0871,
"step": 150
},
{
"epoch": 2.1621621621621623,
"grad_norm": 19.504608154296875,
"learning_rate": 6.27211626663071e-06,
"loss": 0.966,
"step": 160
},
{
"epoch": 2.2972972972972974,
"grad_norm": 22.5753173828125,
"learning_rate": 4.588887071690491e-06,
"loss": 0.9257,
"step": 170
},
{
"epoch": 2.4324324324324325,
"grad_norm": 30.79815101623535,
"learning_rate": 3.12755267356119e-06,
"loss": 0.9889,
"step": 180
},
{
"epoch": 2.5675675675675675,
"grad_norm": 22.710554122924805,
"learning_rate": 1.9192588778428844e-06,
"loss": 0.9634,
"step": 190
},
{
"epoch": 2.7027027027027026,
"grad_norm": 27.529897689819336,
"learning_rate": 9.897583690117813e-07,
"loss": 0.9673,
"step": 200
},
{
"epoch": 2.8378378378378377,
"grad_norm": 23.513914108276367,
"learning_rate": 3.588618366610941e-07,
"loss": 1.0004,
"step": 210
},
{
"epoch": 2.972972972972973,
"grad_norm": 20.257404327392578,
"learning_rate": 4.0015745046725336e-08,
"loss": 0.962,
"step": 220
},
{
"epoch": 3.0,
"step": 222,
"total_flos": 967989855780864.0,
"train_loss": 1.1366610301507485,
"train_runtime": 118.4216,
"train_samples_per_second": 29.792,
"train_steps_per_second": 1.875
}
],
"logging_steps": 10,
"max_steps": 222,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": false,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 967989855780864.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}