| { |
| "best_global_step": 4464, |
| "best_metric": 3.094877004623413, |
| "best_model_checkpoint": null, |
| "epoch": 10.080645161290322, |
| "eval_steps": 500, |
| "global_step": 5000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.0020161290322580645, |
| "grad_norm": 28.535022735595703, |
| "learning_rate": 0.0, |
| "loss": 10.3231, |
| "step": 1 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 4.756303787231445, |
| "learning_rate": 9.900000000000001e-05, |
| "loss": 4.5921, |
| "step": 496 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 3.5128679275512695, |
| "eval_runtime": 19.6704, |
| "eval_samples_per_second": 1618.525, |
| "eval_steps_per_second": 6.355, |
| "step": 496 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 3.945728063583374, |
| "learning_rate": 9.477659574468086e-05, |
| "loss": 3.3197, |
| "step": 992 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 3.296919584274292, |
| "eval_runtime": 19.7075, |
| "eval_samples_per_second": 1615.48, |
| "eval_steps_per_second": 6.343, |
| "step": 992 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 3.4638240337371826, |
| "learning_rate": 8.950000000000001e-05, |
| "loss": 3.1648, |
| "step": 1488 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 3.2211971282958984, |
| "eval_runtime": 19.6423, |
| "eval_samples_per_second": 1620.836, |
| "eval_steps_per_second": 6.364, |
| "step": 1488 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 3.9391467571258545, |
| "learning_rate": 8.422340425531914e-05, |
| "loss": 3.0772, |
| "step": 1984 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 3.1721932888031006, |
| "eval_runtime": 19.9493, |
| "eval_samples_per_second": 1595.896, |
| "eval_steps_per_second": 6.266, |
| "step": 1984 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 4.159926414489746, |
| "learning_rate": 7.894680851063831e-05, |
| "loss": 3.0122, |
| "step": 2480 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 3.142347574234009, |
| "eval_runtime": 19.6986, |
| "eval_samples_per_second": 1616.208, |
| "eval_steps_per_second": 6.346, |
| "step": 2480 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 4.257205009460449, |
| "learning_rate": 7.367021276595744e-05, |
| "loss": 2.9591, |
| "step": 2976 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 3.121657371520996, |
| "eval_runtime": 19.7215, |
| "eval_samples_per_second": 1614.326, |
| "eval_steps_per_second": 6.338, |
| "step": 2976 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 4.3956170082092285, |
| "learning_rate": 6.83936170212766e-05, |
| "loss": 2.9117, |
| "step": 3472 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 3.1076769828796387, |
| "eval_runtime": 19.7076, |
| "eval_samples_per_second": 1615.472, |
| "eval_steps_per_second": 6.343, |
| "step": 3472 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 4.139720916748047, |
| "learning_rate": 6.311702127659575e-05, |
| "loss": 2.8674, |
| "step": 3968 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 3.097724437713623, |
| "eval_runtime": 19.9221, |
| "eval_samples_per_second": 1598.075, |
| "eval_steps_per_second": 6.274, |
| "step": 3968 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 4.987766742706299, |
| "learning_rate": 5.7840425531914896e-05, |
| "loss": 2.8258, |
| "step": 4464 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 3.094877004623413, |
| "eval_runtime": 19.7658, |
| "eval_samples_per_second": 1610.713, |
| "eval_steps_per_second": 6.324, |
| "step": 4464 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 4.529844284057617, |
| "learning_rate": 5.256382978723404e-05, |
| "loss": 2.785, |
| "step": 4960 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 3.0965349674224854, |
| "eval_runtime": 19.7943, |
| "eval_samples_per_second": 1608.392, |
| "eval_steps_per_second": 6.315, |
| "step": 4960 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 9920, |
| "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": 8.344883625984e+16, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
