| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.002109482122139015, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.2189642442780295e-05, | |
| "eval_loss": 11.131023406982422, | |
| "eval_runtime": 10.1897, | |
| "eval_samples_per_second": 489.709, | |
| "eval_steps_per_second": 244.855, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00042189642442780296, | |
| "grad_norm": 2.874541759490967, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 22.1984, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0005484653517561439, | |
| "eval_loss": 10.96646785736084, | |
| "eval_runtime": 9.8347, | |
| "eval_samples_per_second": 507.389, | |
| "eval_steps_per_second": 253.695, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0008437928488556059, | |
| "grad_norm": 2.3113207817077637, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 21.8974, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0010969307035122878, | |
| "eval_loss": 10.829680442810059, | |
| "eval_runtime": 9.8706, | |
| "eval_samples_per_second": 505.543, | |
| "eval_steps_per_second": 252.772, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0012656892732834089, | |
| "grad_norm": 2.328935384750366, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 21.6948, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0016453960552684317, | |
| "eval_loss": 10.776100158691406, | |
| "eval_runtime": 10.0767, | |
| "eval_samples_per_second": 495.201, | |
| "eval_steps_per_second": 247.6, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0016875856977112118, | |
| "grad_norm": 1.9517555236816406, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 21.574, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.002109482122139015, | |
| "grad_norm": 2.224480628967285, | |
| "learning_rate": 0.0, | |
| "loss": 21.5531, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 23396352000.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |