| { |
| "best_global_step": 306, |
| "best_metric": 0.7959078550338745, |
| "best_model_checkpoint": "models/my-synonym-model/checkpoint-306", |
| "epoch": 4.0, |
| "eval_steps": 500, |
| "global_step": 1224, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.6535947712418301, |
| "grad_norm": 20.68070411682129, |
| "learning_rate": 8.548790658882402e-06, |
| "loss": 7.707, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.7959078550338745, |
| "eval_runtime": 0.3262, |
| "eval_samples_per_second": 2507.448, |
| "eval_steps_per_second": 12.261, |
| "step": 306 |
| }, |
| { |
| "epoch": 1.3071895424836601, |
| "grad_norm": 20.929737091064453, |
| "learning_rate": 6.880733944954129e-06, |
| "loss": 6.7597, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.9607843137254903, |
| "grad_norm": 18.126134872436523, |
| "learning_rate": 5.212677231025856e-06, |
| "loss": 6.4126, |
| "step": 600 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 1.0729542970657349, |
| "eval_runtime": 0.4221, |
| "eval_samples_per_second": 1937.794, |
| "eval_steps_per_second": 9.476, |
| "step": 612 |
| }, |
| { |
| "epoch": 2.6143790849673203, |
| "grad_norm": 18.292388916015625, |
| "learning_rate": 3.5446205170975817e-06, |
| "loss": 6.2623, |
| "step": 800 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 1.1428591012954712, |
| "eval_runtime": 0.3576, |
| "eval_samples_per_second": 2287.344, |
| "eval_steps_per_second": 11.185, |
| "step": 918 |
| }, |
| { |
| "epoch": 3.2679738562091503, |
| "grad_norm": 21.801111221313477, |
| "learning_rate": 1.876563803169308e-06, |
| "loss": 6.1651, |
| "step": 1000 |
| }, |
| { |
| "epoch": 3.9215686274509802, |
| "grad_norm": 20.55330467224121, |
| "learning_rate": 2.085070892410342e-07, |
| "loss": 6.12, |
| "step": 1200 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 1.2153398990631104, |
| "eval_runtime": 0.3643, |
| "eval_samples_per_second": 2245.405, |
| "eval_steps_per_second": 10.98, |
| "step": 1224 |
| } |
| ], |
| "logging_steps": 200, |
| "max_steps": 1224, |
| "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": 0.0, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
