| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.08019246190858059, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0016038492381716118, | |
| "eval_loss": NaN, | |
| "eval_runtime": 7.3742, | |
| "eval_samples_per_second": 35.665, | |
| "eval_steps_per_second": 17.9, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00801924619085806, | |
| "grad_norm": 55.09348678588867, | |
| "learning_rate": 5e-05, | |
| "loss": 11.4508, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01603849238171612, | |
| "grad_norm": 40.246456146240234, | |
| "learning_rate": 0.0001, | |
| "loss": 10.4514, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01603849238171612, | |
| "eval_loss": NaN, | |
| "eval_runtime": 5.4461, | |
| "eval_samples_per_second": 48.291, | |
| "eval_steps_per_second": 24.237, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.024057738572574178, | |
| "grad_norm": 42.92169189453125, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 11.0683, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.03207698476343224, | |
| "grad_norm": 44.35247802734375, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 10.8371, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03207698476343224, | |
| "eval_loss": NaN, | |
| "eval_runtime": 5.4284, | |
| "eval_samples_per_second": 48.449, | |
| "eval_steps_per_second": 24.316, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.040096230954290296, | |
| "grad_norm": 37.52597427368164, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 10.8294, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.048115477145148355, | |
| "grad_norm": 36.30201721191406, | |
| "learning_rate": 5e-05, | |
| "loss": 10.9661, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.048115477145148355, | |
| "eval_loss": NaN, | |
| "eval_runtime": 5.3643, | |
| "eval_samples_per_second": 49.028, | |
| "eval_steps_per_second": 24.607, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.056134723336006415, | |
| "grad_norm": 35.2260856628418, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 11.3404, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.06415396952686447, | |
| "grad_norm": 40.70652770996094, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 11.0508, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.06415396952686447, | |
| "eval_loss": NaN, | |
| "eval_runtime": 5.5162, | |
| "eval_samples_per_second": 47.678, | |
| "eval_steps_per_second": 23.929, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07217321571772253, | |
| "grad_norm": 44.69584274291992, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 10.269, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.08019246190858059, | |
| "grad_norm": 63.653263092041016, | |
| "learning_rate": 0.0, | |
| "loss": 10.6362, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08019246190858059, | |
| "eval_loss": NaN, | |
| "eval_runtime": 5.3076, | |
| "eval_samples_per_second": 49.552, | |
| "eval_steps_per_second": 24.87, | |
| "step": 50 | |
| } | |
| ], | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 449608899624960.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |