| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.45558086560364464, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.009111617312072893, | |
| "eval_loss": 13.944470405578613, | |
| "eval_runtime": 28.8071, | |
| "eval_samples_per_second": 25.688, | |
| "eval_steps_per_second": 3.228, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.09111617312072894, | |
| "grad_norm": 8.145538330078125, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 12.18, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.09111617312072894, | |
| "eval_loss": 8.518851280212402, | |
| "eval_runtime": 27.4623, | |
| "eval_samples_per_second": 26.946, | |
| "eval_steps_per_second": 3.386, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.18223234624145787, | |
| "grad_norm": 12.082337379455566, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 5.7398, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.18223234624145787, | |
| "eval_loss": 2.1542670726776123, | |
| "eval_runtime": 27.6859, | |
| "eval_samples_per_second": 26.728, | |
| "eval_steps_per_second": 3.359, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2733485193621868, | |
| "grad_norm": 2.113602876663208, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 1.0209, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2733485193621868, | |
| "eval_loss": 0.4289436638355255, | |
| "eval_runtime": 27.4522, | |
| "eval_samples_per_second": 26.956, | |
| "eval_steps_per_second": 3.388, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.36446469248291574, | |
| "grad_norm": 2.657161235809326, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 0.4525, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.36446469248291574, | |
| "eval_loss": 0.3867056369781494, | |
| "eval_runtime": 27.7318, | |
| "eval_samples_per_second": 26.684, | |
| "eval_steps_per_second": 3.354, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.45558086560364464, | |
| "grad_norm": 0.9272085428237915, | |
| "learning_rate": 0.0, | |
| "loss": 0.3723, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.45558086560364464, | |
| "eval_loss": 0.37940335273742676, | |
| "eval_runtime": 27.5173, | |
| "eval_samples_per_second": 26.892, | |
| "eval_steps_per_second": 3.38, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.136884414860493e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |