rak-r05's picture
Training in progress, step 10, checkpoint
1fa964a verified
raw
history blame
2.95 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.021941854086670324,
"eval_steps": 3,
"global_step": 10,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0021941854086670325,
"grad_norm": NaN,
"learning_rate": 2e-05,
"loss": 0.0,
"step": 1
},
{
"epoch": 0.0021941854086670325,
"eval_loss": NaN,
"eval_runtime": 7.1045,
"eval_samples_per_second": 27.025,
"eval_steps_per_second": 13.513,
"step": 1
},
{
"epoch": 0.004388370817334065,
"grad_norm": NaN,
"learning_rate": 4e-05,
"loss": 0.0,
"step": 2
},
{
"epoch": 0.006582556226001097,
"grad_norm": NaN,
"learning_rate": 6e-05,
"loss": 0.0,
"step": 3
},
{
"epoch": 0.006582556226001097,
"eval_loss": NaN,
"eval_runtime": 5.5703,
"eval_samples_per_second": 34.469,
"eval_steps_per_second": 17.234,
"step": 3
},
{
"epoch": 0.00877674163466813,
"grad_norm": NaN,
"learning_rate": 8e-05,
"loss": 0.2507,
"step": 4
},
{
"epoch": 0.010970927043335162,
"grad_norm": NaN,
"learning_rate": 0.0001,
"loss": 0.0,
"step": 5
},
{
"epoch": 0.013165112452002194,
"grad_norm": NaN,
"learning_rate": 0.00012,
"loss": 0.0,
"step": 6
},
{
"epoch": 0.013165112452002194,
"eval_loss": NaN,
"eval_runtime": 5.5928,
"eval_samples_per_second": 34.33,
"eval_steps_per_second": 17.165,
"step": 6
},
{
"epoch": 0.015359297860669226,
"grad_norm": NaN,
"learning_rate": 0.00014,
"loss": 0.0,
"step": 7
},
{
"epoch": 0.01755348326933626,
"grad_norm": NaN,
"learning_rate": 0.00016,
"loss": 0.0,
"step": 8
},
{
"epoch": 0.019747668678003292,
"grad_norm": NaN,
"learning_rate": 0.00018,
"loss": 0.0,
"step": 9
},
{
"epoch": 0.019747668678003292,
"eval_loss": NaN,
"eval_runtime": 5.6133,
"eval_samples_per_second": 34.205,
"eval_steps_per_second": 17.102,
"step": 9
},
{
"epoch": 0.021941854086670324,
"grad_norm": NaN,
"learning_rate": 0.0002,
"loss": 0.0,
"step": 10
}
],
"logging_steps": 1,
"max_steps": 10,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 3,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 330750083727360.0,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}