phi-2-gpo-test-iter-0 / trainer_state.json
BraylonDash's picture
Model save
0e6ceac verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.0,
"eval_steps": 100,
"global_step": 2,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"learning_rate": 5e-06,
"logits/chosen": 0.7793408036231995,
"logits/rejected": 0.853427529335022,
"logps/chosen": -255.33981323242188,
"logps/rejected": -306.3643798828125,
"loss": 0.0001,
"rewards/accuracies": 0.0,
"rewards/chosen": 0.0,
"rewards/margins": 0.0,
"rewards/rejected": 0.0,
"step": 1
},
{
"epoch": 2.0,
"step": 2,
"total_flos": 0.0,
"train_loss": 0.00012263256940059364,
"train_runtime": 3.7437,
"train_samples_per_second": 2.137,
"train_steps_per_second": 0.534
}
],
"logging_steps": 10,
"max_steps": 2,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 100,
"total_flos": 0.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}