| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 10.0, |
| "eval_steps": 500, |
| "global_step": 8650, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.18263820939334638, |
| "eval_loss": 5.924222469329834, |
| "eval_runtime": 4.5576, |
| "eval_samples_per_second": 42.127, |
| "eval_steps_per_second": 1.316, |
| "step": 865 |
| }, |
| { |
| "epoch": 1.1560693641618498, |
| "grad_norm": 10577.5283203125, |
| "learning_rate": 0.0005993999999999999, |
| "loss": 1.4825, |
| "step": 1000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.2094035388127854, |
| "eval_loss": 5.4699883460998535, |
| "eval_runtime": 4.2355, |
| "eval_samples_per_second": 45.331, |
| "eval_steps_per_second": 1.417, |
| "step": 1730 |
| }, |
| { |
| "epoch": 2.3121387283236996, |
| "grad_norm": 10539.6064453125, |
| "learning_rate": 0.0005216470588235294, |
| "loss": 1.278, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.2329786366601435, |
| "eval_loss": 5.17898416519165, |
| "eval_runtime": 4.1051, |
| "eval_samples_per_second": 46.771, |
| "eval_steps_per_second": 1.462, |
| "step": 2595 |
| }, |
| { |
| "epoch": 3.468208092485549, |
| "grad_norm": 10572.8818359375, |
| "learning_rate": 0.00044321568627450977, |
| "loss": 1.1768, |
| "step": 3000 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.24628995433789955, |
| "eval_loss": 5.025396823883057, |
| "eval_runtime": 4.1012, |
| "eval_samples_per_second": 46.815, |
| "eval_steps_per_second": 1.463, |
| "step": 3460 |
| }, |
| { |
| "epoch": 4.624277456647399, |
| "grad_norm": 11170.2646484375, |
| "learning_rate": 0.0003647843137254902, |
| "loss": 1.1207, |
| "step": 4000 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.2575729778212655, |
| "eval_loss": 4.892265319824219, |
| "eval_runtime": 4.0658, |
| "eval_samples_per_second": 47.223, |
| "eval_steps_per_second": 1.476, |
| "step": 4325 |
| }, |
| { |
| "epoch": 5.780346820809249, |
| "grad_norm": 11376.833984375, |
| "learning_rate": 0.00028635294117647055, |
| "loss": 1.083, |
| "step": 5000 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.26632827788649704, |
| "eval_loss": 4.8085503578186035, |
| "eval_runtime": 4.1174, |
| "eval_samples_per_second": 46.631, |
| "eval_steps_per_second": 1.457, |
| "step": 5190 |
| }, |
| { |
| "epoch": 6.936416184971098, |
| "grad_norm": 11552.1865234375, |
| "learning_rate": 0.00020792156862745097, |
| "loss": 1.0491, |
| "step": 6000 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.2750224233529028, |
| "eval_loss": 4.724864482879639, |
| "eval_runtime": 4.0931, |
| "eval_samples_per_second": 46.908, |
| "eval_steps_per_second": 1.466, |
| "step": 6055 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.2814028864970646, |
| "eval_loss": 4.664273738861084, |
| "eval_runtime": 4.1968, |
| "eval_samples_per_second": 45.749, |
| "eval_steps_per_second": 1.43, |
| "step": 6920 |
| }, |
| { |
| "epoch": 8.092485549132949, |
| "grad_norm": 12136.4814453125, |
| "learning_rate": 0.00012949019607843136, |
| "loss": 1.0193, |
| "step": 7000 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.2878954664057404, |
| "eval_loss": 4.605855464935303, |
| "eval_runtime": 4.1869, |
| "eval_samples_per_second": 45.858, |
| "eval_steps_per_second": 1.433, |
| "step": 7785 |
| }, |
| { |
| "epoch": 9.248554913294798, |
| "grad_norm": 12644.693359375, |
| "learning_rate": 5.105882352941176e-05, |
| "loss": 0.9915, |
| "step": 8000 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_accuracy": 0.2907289628180039, |
| "eval_loss": 4.5860066413879395, |
| "eval_runtime": 4.1305, |
| "eval_samples_per_second": 46.483, |
| "eval_steps_per_second": 1.453, |
| "step": 8650 |
| }, |
| { |
| "epoch": 10.0, |
| "step": 8650, |
| "total_flos": 7.227860189184e+16, |
| "train_loss": 1.1371644900988982, |
| "train_runtime": 8936.9854, |
| "train_samples_per_second": 30.952, |
| "train_steps_per_second": 0.968 |
| } |
| ], |
| "logging_steps": 1000, |
| "max_steps": 8650, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 10, |
| "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": 7.227860189184e+16, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
