| { |
| "best_global_step": 1984, |
| "best_metric": 2.807705879211426, |
| "best_model_checkpoint": null, |
| "epoch": 4.032258064516129, |
| "eval_steps": 500, |
| "global_step": 2000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.0020161290322580645, |
| "grad_norm": 29.451557159423828, |
| "learning_rate": 0.0, |
| "loss": 10.3668, |
| "step": 1 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.9898180961608887, |
| "learning_rate": 9.900000000000001e-05, |
| "loss": 4.4948, |
| "step": 496 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 3.2884881496429443, |
| "eval_runtime": 19.6318, |
| "eval_samples_per_second": 1622.471, |
| "eval_steps_per_second": 6.367, |
| "step": 496 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.733898401260376, |
| "learning_rate": 9.477659574468086e-05, |
| "loss": 3.0395, |
| "step": 992 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 2.9790470600128174, |
| "eval_runtime": 19.6746, |
| "eval_samples_per_second": 1618.936, |
| "eval_steps_per_second": 6.353, |
| "step": 992 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.7452684640884399, |
| "learning_rate": 8.950000000000001e-05, |
| "loss": 2.8091, |
| "step": 1488 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 2.8684024810791016, |
| "eval_runtime": 19.5669, |
| "eval_samples_per_second": 1627.847, |
| "eval_steps_per_second": 6.388, |
| "step": 1488 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.7501894235610962, |
| "learning_rate": 8.422340425531914e-05, |
| "loss": 2.6886, |
| "step": 1984 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 2.807705879211426, |
| "eval_runtime": 19.9207, |
| "eval_samples_per_second": 1598.937, |
| "eval_steps_per_second": 6.275, |
| "step": 1984 |
| } |
| ], |
| "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": 3.3426567069696e+16, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
