| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03073140749846343, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0006146281499692685, | |
| "eval_loss": 1.4701268672943115, | |
| "eval_runtime": 69.7203, | |
| "eval_samples_per_second": 9.825, | |
| "eval_steps_per_second": 4.92, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.003073140749846343, | |
| "grad_norm": 0.4191581606864929, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0406, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.006146281499692686, | |
| "grad_norm": 0.4920947551727295, | |
| "learning_rate": 0.0001, | |
| "loss": 1.1383, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.006146281499692686, | |
| "eval_loss": 1.4146133661270142, | |
| "eval_runtime": 70.4441, | |
| "eval_samples_per_second": 9.724, | |
| "eval_steps_per_second": 4.869, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00921942224953903, | |
| "grad_norm": 0.5874287486076355, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.9912, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.012292562999385371, | |
| "grad_norm": 0.8421012759208679, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.8763, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.012292562999385371, | |
| "eval_loss": 1.1583969593048096, | |
| "eval_runtime": 70.5715, | |
| "eval_samples_per_second": 9.706, | |
| "eval_steps_per_second": 4.86, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.015365703749231715, | |
| "grad_norm": 0.6995245814323425, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.0697, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01843884449907806, | |
| "grad_norm": 0.8031644225120544, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0365, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.01843884449907806, | |
| "eval_loss": 1.047980546951294, | |
| "eval_runtime": 70.4388, | |
| "eval_samples_per_second": 9.725, | |
| "eval_steps_per_second": 4.869, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0215119852489244, | |
| "grad_norm": 0.8080487251281738, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.1517, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.024585125998770743, | |
| "grad_norm": 0.908812940120697, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.9924, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.024585125998770743, | |
| "eval_loss": 1.0201315879821777, | |
| "eval_runtime": 70.5948, | |
| "eval_samples_per_second": 9.703, | |
| "eval_steps_per_second": 4.859, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.027658266748617086, | |
| "grad_norm": 0.6947146058082581, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.9797, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.03073140749846343, | |
| "grad_norm": 0.7228668928146362, | |
| "learning_rate": 0.0, | |
| "loss": 1.081, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.03073140749846343, | |
| "eval_loss": 1.0150693655014038, | |
| "eval_runtime": 70.601, | |
| "eval_samples_per_second": 9.702, | |
| "eval_steps_per_second": 4.858, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "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": 1.4935231303778304e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |