| { |
| "best_global_step": 4788, |
| "best_metric": 5.561720848083496, |
| "best_model_checkpoint": null, |
| "epoch": 12.531328320802006, |
| "eval_steps": 500, |
| "global_step": 5000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.002506265664160401, |
| "grad_norm": 16.206899642944336, |
| "learning_rate": 0.0, |
| "loss": 10.3203, |
| "step": 1 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.7419695258140564, |
| "learning_rate": 7.960000000000001e-05, |
| "loss": 7.2424, |
| "step": 399 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 6.557571887969971, |
| "eval_runtime": 15.7818, |
| "eval_samples_per_second": 1621.551, |
| "eval_steps_per_second": 6.336, |
| "step": 399 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.8143869042396545, |
| "learning_rate": 9.601876675603218e-05, |
| "loss": 6.3077, |
| "step": 798 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 6.204631328582764, |
| "eval_runtime": 15.7933, |
| "eval_samples_per_second": 1620.371, |
| "eval_steps_per_second": 6.332, |
| "step": 798 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.8700923323631287, |
| "learning_rate": 9.067024128686328e-05, |
| "loss": 6.0417, |
| "step": 1197 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 6.021187782287598, |
| "eval_runtime": 15.8286, |
| "eval_samples_per_second": 1616.753, |
| "eval_steps_per_second": 6.318, |
| "step": 1197 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.9790167212486267, |
| "learning_rate": 8.532171581769438e-05, |
| "loss": 5.8739, |
| "step": 1596 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 5.898504257202148, |
| "eval_runtime": 15.7952, |
| "eval_samples_per_second": 1620.176, |
| "eval_steps_per_second": 6.331, |
| "step": 1596 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 1.0000659227371216, |
| "learning_rate": 7.997319034852548e-05, |
| "loss": 5.7396, |
| "step": 1995 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 5.8024163246154785, |
| "eval_runtime": 15.8237, |
| "eval_samples_per_second": 1617.262, |
| "eval_steps_per_second": 6.32, |
| "step": 1995 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 0.9891586303710938, |
| "learning_rate": 7.462466487935658e-05, |
| "loss": 5.6246, |
| "step": 2394 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 5.729940414428711, |
| "eval_runtime": 15.8227, |
| "eval_samples_per_second": 1617.36, |
| "eval_steps_per_second": 6.32, |
| "step": 2394 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 1.0765498876571655, |
| "learning_rate": 6.927613941018766e-05, |
| "loss": 5.5237, |
| "step": 2793 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 5.674533367156982, |
| "eval_runtime": 15.8362, |
| "eval_samples_per_second": 1615.977, |
| "eval_steps_per_second": 6.315, |
| "step": 2793 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 1.1928595304489136, |
| "learning_rate": 6.392761394101878e-05, |
| "loss": 5.4334, |
| "step": 3192 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 5.6343817710876465, |
| "eval_runtime": 15.8433, |
| "eval_samples_per_second": 1615.256, |
| "eval_steps_per_second": 6.312, |
| "step": 3192 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 1.2139155864715576, |
| "learning_rate": 5.8579088471849864e-05, |
| "loss": 5.3507, |
| "step": 3591 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 5.601651191711426, |
| "eval_runtime": 15.8447, |
| "eval_samples_per_second": 1615.111, |
| "eval_steps_per_second": 6.311, |
| "step": 3591 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 1.3513448238372803, |
| "learning_rate": 5.3230563002680965e-05, |
| "loss": 5.2753, |
| "step": 3990 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 5.58331298828125, |
| "eval_runtime": 15.8245, |
| "eval_samples_per_second": 1617.173, |
| "eval_steps_per_second": 6.319, |
| "step": 3990 |
| }, |
| { |
| "epoch": 11.0, |
| "grad_norm": 1.4445401430130005, |
| "learning_rate": 4.7882037533512065e-05, |
| "loss": 5.2045, |
| "step": 4389 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_loss": 5.569336891174316, |
| "eval_runtime": 15.8642, |
| "eval_samples_per_second": 1613.124, |
| "eval_steps_per_second": 6.303, |
| "step": 4389 |
| }, |
| { |
| "epoch": 12.0, |
| "grad_norm": 1.5078763961791992, |
| "learning_rate": 4.2533512064343165e-05, |
| "loss": 5.1388, |
| "step": 4788 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_loss": 5.561720848083496, |
| "eval_runtime": 15.8463, |
| "eval_samples_per_second": 1614.949, |
| "eval_steps_per_second": 6.311, |
| "step": 4788 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 7980, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 20, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 8.3487246188544e+16, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
