| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.909090909090909, | |
| "eval_steps": 2, | |
| "global_step": 16, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1.0163, | |
| "eval_samples_per_second": 4.92, | |
| "eval_steps_per_second": 0.984, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.5198, | |
| "eval_samples_per_second": 9.619, | |
| "eval_steps_per_second": 1.924, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.5199, | |
| "eval_samples_per_second": 9.618, | |
| "eval_steps_per_second": 1.924, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.5207, | |
| "eval_samples_per_second": 9.603, | |
| "eval_steps_per_second": 1.921, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.2727272727272727, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.5211, | |
| "eval_samples_per_second": 9.594, | |
| "eval_steps_per_second": 1.919, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.6363636363636362, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.5202, | |
| "eval_samples_per_second": 9.611, | |
| "eval_steps_per_second": 1.922, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.5221, | |
| "eval_samples_per_second": 9.577, | |
| "eval_steps_per_second": 1.915, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.3636363636363638, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.522, | |
| "eval_samples_per_second": 9.579, | |
| "eval_steps_per_second": 1.916, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.5202, | |
| "eval_samples_per_second": 9.611, | |
| "eval_steps_per_second": 1.922, | |
| "step": 16 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 16, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0922479583232e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |