| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.006117287218398918, | |
| "eval_steps": 10, | |
| "global_step": 26, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0002352802776307276, | |
| "eval_loss": 3.421792507171631, | |
| "eval_runtime": 125.7544, | |
| "eval_samples_per_second": 14.234, | |
| "eval_steps_per_second": 7.117, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.001176401388153638, | |
| "grad_norm": 1.2427492141723633, | |
| "learning_rate": 5e-05, | |
| "loss": 3.4053, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.002352802776307276, | |
| "grad_norm": 2.223381280899048, | |
| "learning_rate": 0.0001, | |
| "loss": 3.4449, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.002352802776307276, | |
| "eval_loss": 3.1467432975769043, | |
| "eval_runtime": 131.4609, | |
| "eval_samples_per_second": 13.616, | |
| "eval_steps_per_second": 6.808, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0035292041644609143, | |
| "grad_norm": 2.0198276042938232, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.8499, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.004705605552614552, | |
| "grad_norm": 3.1428215503692627, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 2.0307, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.004705605552614552, | |
| "eval_loss": 1.6161822080612183, | |
| "eval_runtime": 131.4252, | |
| "eval_samples_per_second": 13.62, | |
| "eval_steps_per_second": 6.81, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00588200694076819, | |
| "grad_norm": 1.708250641822815, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.2073, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4272226987671552.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |