deepseek_setting / checkpoint-3 /trainer_state.json
Kei5uke's picture
Training in progress, step 3, checkpoint
2f9599c verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.0018601767167880949,
"eval_steps": 3,
"global_step": 3,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0006200589055960316,
"grad_norm": 0.6006524562835693,
"learning_rate": 4.000000000000001e-06,
"loss": 1.4363,
"step": 1
},
{
"epoch": 0.0012401178111920632,
"grad_norm": 0.4452102482318878,
"learning_rate": 8.000000000000001e-06,
"loss": 1.1641,
"step": 2
},
{
"epoch": 0.0018601767167880949,
"grad_norm": 0.418393075466156,
"learning_rate": 1.2e-05,
"loss": 1.0005,
"step": 3
},
{
"epoch": 0.0018601767167880949,
"eval_loss": 1.2559138536453247,
"eval_runtime": 50.2632,
"eval_samples_per_second": 1.99,
"eval_steps_per_second": 1.99,
"step": 3
}
],
"logging_steps": 1,
"max_steps": 30,
"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": false
},
"attributes": {}
}
},
"total_flos": 7009626086117376.0,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}