| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.0, |
| "eval_steps": 2000, |
| "global_step": 10710, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.0001868460388639761, |
| "grad_norm": 4.329383373260498, |
| "learning_rate": 0.0, |
| "loss": 2.3475, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.37369207772795215, |
| "grad_norm": 1.3929723501205444, |
| "learning_rate": 4.6683792620270904e-05, |
| "loss": 1.4174, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.37369207772795215, |
| "eval_cosine_accuracy": 0.9179443717002869, |
| "eval_loss": 0.44630110263824463, |
| "eval_runtime": 28.0643, |
| "eval_samples_per_second": 336.976, |
| "eval_steps_per_second": 1.318, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.7473841554559043, |
| "grad_norm": 1.1317882537841797, |
| "learning_rate": 4.886145894517875e-05, |
| "loss": 0.7274, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.7473841554559043, |
| "eval_cosine_accuracy": 0.9551655054092407, |
| "eval_loss": 0.38517987728118896, |
| "eval_runtime": 27.0397, |
| "eval_samples_per_second": 349.745, |
| "eval_steps_per_second": 1.368, |
| "step": 4000 |
| }, |
| { |
| "epoch": 1.1209406494960805, |
| "grad_norm": 0.9535921812057495, |
| "learning_rate": 4.521439737687342e-05, |
| "loss": 0.5395, |
| "step": 6000 |
| }, |
| { |
| "epoch": 1.1209406494960805, |
| "eval_cosine_accuracy": 0.9591836929321289, |
| "eval_loss": 0.35122039914131165, |
| "eval_runtime": 27.55, |
| "eval_samples_per_second": 343.266, |
| "eval_steps_per_second": 1.343, |
| "step": 6000 |
| }, |
| { |
| "epoch": 1.494214259051885, |
| "grad_norm": 1.612464427947998, |
| "learning_rate": 3.94365487916572e-05, |
| "loss": 0.4559, |
| "step": 8000 |
| }, |
| { |
| "epoch": 1.494214259051885, |
| "eval_cosine_accuracy": 0.9590779542922974, |
| "eval_loss": 0.3402463495731354, |
| "eval_runtime": 27.2192, |
| "eval_samples_per_second": 347.439, |
| "eval_steps_per_second": 1.359, |
| "step": 8000 |
| }, |
| { |
| "epoch": 1.8674878686076894, |
| "grad_norm": 1.5486053228378296, |
| "learning_rate": 3.214085926202344e-05, |
| "loss": 0.4172, |
| "step": 10000 |
| }, |
| { |
| "epoch": 1.8674878686076894, |
| "eval_cosine_accuracy": 0.9661626219749451, |
| "eval_loss": 0.3279242217540741, |
| "eval_runtime": 27.4211, |
| "eval_samples_per_second": 344.88, |
| "eval_steps_per_second": 1.349, |
| "step": 10000 |
| } |
| ], |
| "logging_steps": 2000, |
| "max_steps": 21408, |
| "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": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
