| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 4.0, |
| "eval_steps": 500, |
| "global_step": 28, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.14285714285714285, |
| "grad_norm": 0.029374502948890973, |
| "learning_rate": 1.6666666666666667e-05, |
| "loss": 0.2105, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.2857142857142857, |
| "grad_norm": 0.02879271003469766, |
| "learning_rate": 3.3333333333333335e-05, |
| "loss": 0.2059, |
| "step": 2 |
| }, |
| { |
| "epoch": 0.42857142857142855, |
| "grad_norm": 0.028417140992448447, |
| "learning_rate": 5e-05, |
| "loss": 0.2136, |
| "step": 3 |
| }, |
| { |
| "epoch": 0.5714285714285714, |
| "grad_norm": 0.029508379259060556, |
| "learning_rate": 4.980286753286195e-05, |
| "loss": 0.2064, |
| "step": 4 |
| }, |
| { |
| "epoch": 0.7142857142857143, |
| "grad_norm": 0.031023789828790938, |
| "learning_rate": 4.9214579028215776e-05, |
| "loss": 0.2051, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.8571428571428571, |
| "grad_norm": 0.03173373208486436, |
| "learning_rate": 4.8244412147206284e-05, |
| "loss": 0.2009, |
| "step": 6 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.02891593307804275, |
| "learning_rate": 4.690766700109659e-05, |
| "loss": 0.1928, |
| "step": 7 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.17100295424461365, |
| "eval_runtime": 2.5986, |
| "eval_samples_per_second": 0.385, |
| "eval_steps_per_second": 0.385, |
| "step": 7 |
| }, |
| { |
| "epoch": 1.1428571428571428, |
| "grad_norm": 0.0291759534514554, |
| "learning_rate": 4.522542485937369e-05, |
| "loss": 0.1816, |
| "step": 8 |
| }, |
| { |
| "epoch": 1.2857142857142856, |
| "grad_norm": 0.03503069911331975, |
| "learning_rate": 4.3224215685535294e-05, |
| "loss": 0.1771, |
| "step": 9 |
| }, |
| { |
| "epoch": 1.4285714285714286, |
| "grad_norm": 0.03222295532632566, |
| "learning_rate": 4.093559974371725e-05, |
| "loss": 0.1714, |
| "step": 10 |
| }, |
| { |
| "epoch": 1.5714285714285714, |
| "grad_norm": 0.02924500419270857, |
| "learning_rate": 3.8395669874474915e-05, |
| "loss": 0.176, |
| "step": 11 |
| }, |
| { |
| "epoch": 1.7142857142857144, |
| "grad_norm": 0.025032804888512103, |
| "learning_rate": 3.564448228912682e-05, |
| "loss": 0.1688, |
| "step": 12 |
| }, |
| { |
| "epoch": 1.8571428571428572, |
| "grad_norm": 0.02333745520711948, |
| "learning_rate": 3.272542485937369e-05, |
| "loss": 0.1641, |
| "step": 13 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.02299528178586051, |
| "learning_rate": 2.9684532864643122e-05, |
| "loss": 0.1516, |
| "step": 14 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.14068603515625, |
| "eval_runtime": 2.4437, |
| "eval_samples_per_second": 0.409, |
| "eval_steps_per_second": 0.409, |
| "step": 14 |
| }, |
| { |
| "epoch": 2.142857142857143, |
| "grad_norm": 0.02266916678811667, |
| "learning_rate": 2.656976298823284e-05, |
| "loss": 0.1553, |
| "step": 15 |
| }, |
| { |
| "epoch": 2.2857142857142856, |
| "grad_norm": 0.02244871211073816, |
| "learning_rate": 2.3430237011767167e-05, |
| "loss": 0.1484, |
| "step": 16 |
| }, |
| { |
| "epoch": 2.4285714285714284, |
| "grad_norm": 0.024630119269621335, |
| "learning_rate": 2.031546713535688e-05, |
| "loss": 0.1489, |
| "step": 17 |
| }, |
| { |
| "epoch": 2.571428571428571, |
| "grad_norm": 0.02310463707931144, |
| "learning_rate": 1.7274575140626318e-05, |
| "loss": 0.1433, |
| "step": 18 |
| }, |
| { |
| "epoch": 2.7142857142857144, |
| "grad_norm": 0.022565531397988565, |
| "learning_rate": 1.4355517710873184e-05, |
| "loss": 0.145, |
| "step": 19 |
| }, |
| { |
| "epoch": 2.857142857142857, |
| "grad_norm": 0.02303743715249304, |
| "learning_rate": 1.1604330125525079e-05, |
| "loss": 0.1388, |
| "step": 20 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.02213833372976756, |
| "learning_rate": 9.064400256282757e-06, |
| "loss": 0.1408, |
| "step": 21 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.12390902638435364, |
| "eval_runtime": 2.5093, |
| "eval_samples_per_second": 0.399, |
| "eval_steps_per_second": 0.399, |
| "step": 21 |
| }, |
| { |
| "epoch": 3.142857142857143, |
| "grad_norm": 0.02206988787856512, |
| "learning_rate": 6.775784314464717e-06, |
| "loss": 0.1386, |
| "step": 22 |
| }, |
| { |
| "epoch": 3.2857142857142856, |
| "grad_norm": 0.02137909507738522, |
| "learning_rate": 4.7745751406263165e-06, |
| "loss": 0.1363, |
| "step": 23 |
| }, |
| { |
| "epoch": 3.4285714285714284, |
| "grad_norm": 0.021550899474365476, |
| "learning_rate": 3.092332998903416e-06, |
| "loss": 0.1256, |
| "step": 24 |
| }, |
| { |
| "epoch": 3.571428571428571, |
| "grad_norm": 0.021209476107635006, |
| "learning_rate": 1.7555878527937164e-06, |
| "loss": 0.1366, |
| "step": 25 |
| }, |
| { |
| "epoch": 3.7142857142857144, |
| "grad_norm": 0.02118940295971805, |
| "learning_rate": 7.854209717842231e-07, |
| "loss": 0.1325, |
| "step": 26 |
| }, |
| { |
| "epoch": 3.857142857142857, |
| "grad_norm": 0.02088779986332402, |
| "learning_rate": 1.9713246713805588e-07, |
| "loss": 0.1335, |
| "step": 27 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.021082645298901994, |
| "learning_rate": 0.0, |
| "loss": 0.1367, |
| "step": 28 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.12003877758979797, |
| "eval_runtime": 2.4052, |
| "eval_samples_per_second": 0.416, |
| "eval_steps_per_second": 0.416, |
| "step": 28 |
| }, |
| { |
| "epoch": 4.0, |
| "step": 28, |
| "total_flos": 613369774080000.0, |
| "train_loss": 0.1637916559619563, |
| "train_runtime": 908.7009, |
| "train_samples_per_second": 0.88, |
| "train_steps_per_second": 0.031 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 28, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 613369774080000.0, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
