142 lines
3.2 KiB
JSON
142 lines
3.2 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 4.0,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 144,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 0.28169014084507044,
|
||
|
|
"grad_norm": 10.5625,
|
||
|
|
"learning_rate": 9.375000000000001e-06,
|
||
|
|
"loss": 2.1714,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5633802816901409,
|
||
|
|
"grad_norm": 8.5,
|
||
|
|
"learning_rate": 8.680555555555557e-06,
|
||
|
|
"loss": 1.3165,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8450704225352113,
|
||
|
|
"grad_norm": 8.75,
|
||
|
|
"learning_rate": 7.986111111111112e-06,
|
||
|
|
"loss": 1.2261,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.1126760563380282,
|
||
|
|
"grad_norm": 6.40625,
|
||
|
|
"learning_rate": 7.291666666666667e-06,
|
||
|
|
"loss": 1.1755,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.3943661971830985,
|
||
|
|
"grad_norm": 6.875,
|
||
|
|
"learning_rate": 6.597222222222223e-06,
|
||
|
|
"loss": 1.0993,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.676056338028169,
|
||
|
|
"grad_norm": 7.78125,
|
||
|
|
"learning_rate": 5.9027777777777785e-06,
|
||
|
|
"loss": 1.0708,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.9577464788732395,
|
||
|
|
"grad_norm": 8.9375,
|
||
|
|
"learning_rate": 5.208333333333334e-06,
|
||
|
|
"loss": 1.0251,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.2253521126760565,
|
||
|
|
"grad_norm": 10.125,
|
||
|
|
"learning_rate": 4.5138888888888895e-06,
|
||
|
|
"loss": 0.9767,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.507042253521127,
|
||
|
|
"grad_norm": 11.4375,
|
||
|
|
"learning_rate": 3.819444444444444e-06,
|
||
|
|
"loss": 0.9844,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.788732394366197,
|
||
|
|
"grad_norm": 13.25,
|
||
|
|
"learning_rate": 3.125e-06,
|
||
|
|
"loss": 0.9411,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.056338028169014,
|
||
|
|
"grad_norm": 13.8125,
|
||
|
|
"learning_rate": 2.4305555555555557e-06,
|
||
|
|
"loss": 0.9409,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.3380281690140845,
|
||
|
|
"grad_norm": 13.375,
|
||
|
|
"learning_rate": 1.7361111111111112e-06,
|
||
|
|
"loss": 0.9274,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.619718309859155,
|
||
|
|
"grad_norm": 14.25,
|
||
|
|
"learning_rate": 1.0416666666666667e-06,
|
||
|
|
"loss": 0.9108,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 3.9014084507042255,
|
||
|
|
"grad_norm": 13.0625,
|
||
|
|
"learning_rate": 3.472222222222223e-07,
|
||
|
|
"loss": 0.9016,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 4.0,
|
||
|
|
"step": 144,
|
||
|
|
"total_flos": 1.967368419724493e+16,
|
||
|
|
"train_loss": 1.1122563150193956,
|
||
|
|
"train_runtime": 255.0306,
|
||
|
|
"train_samples_per_second": 8.94,
|
||
|
|
"train_steps_per_second": 0.565
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 10,
|
||
|
|
"max_steps": 144,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 4,
|
||
|
|
"save_steps": 35,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": true
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 1.967368419724493e+16,
|
||
|
|
"train_batch_size": 8,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|