| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0005589090096132349, | |
| "eval_steps": 25, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.11781801922647e-05, | |
| "eval_loss": 2.3351893424987793, | |
| "eval_runtime": 2611.9375, | |
| "eval_samples_per_second": 3.806, | |
| "eval_steps_per_second": 3.806, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 5.58909009613235e-05, | |
| "grad_norm": 8.791929244995117, | |
| "learning_rate": 5e-05, | |
| "loss": 1.203, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.000111781801922647, | |
| "grad_norm": 3.972585916519165, | |
| "learning_rate": 0.0001, | |
| "loss": 2.1104, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00016767270288397049, | |
| "grad_norm": 11.146024703979492, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 1.2193, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.000223563603845294, | |
| "grad_norm": 1.1209495067596436, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.7004, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00027945450480661746, | |
| "grad_norm": 8.159049987792969, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.6708, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.00027945450480661746, | |
| "eval_loss": 0.5174654126167297, | |
| "eval_runtime": 2611.3449, | |
| "eval_samples_per_second": 3.806, | |
| "eval_steps_per_second": 3.806, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.00033534540576794097, | |
| "grad_norm": 8.198873519897461, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4893, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0003912363067292645, | |
| "grad_norm": 3.9584426879882812, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.6039, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.000447127207690588, | |
| "grad_norm": 3.658045768737793, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.8857, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0005030181086519115, | |
| "grad_norm": 3.1578454971313477, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.4418, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0005589090096132349, | |
| "grad_norm": 6.337815761566162, | |
| "learning_rate": 0.0, | |
| "loss": 0.3068, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0005589090096132349, | |
| "eval_loss": 0.42810094356536865, | |
| "eval_runtime": 2596.8666, | |
| "eval_samples_per_second": 3.828, | |
| "eval_steps_per_second": 3.828, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1851583440617472.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |