| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0032928759894459104, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.443271767810026e-05, | |
| "eval_loss": 6.900400638580322, | |
| "eval_runtime": 415.9031, | |
| "eval_samples_per_second": 11.991, | |
| "eval_steps_per_second": 5.997, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00042216358839050134, | |
| "grad_norm": 225.8194580078125, | |
| "learning_rate": 5e-05, | |
| "loss": 26.3712, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0008443271767810027, | |
| "grad_norm": 205.8764190673828, | |
| "learning_rate": 0.0001, | |
| "loss": 10.7889, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0008443271767810027, | |
| "eval_loss": 1.4552971124649048, | |
| "eval_runtime": 425.3753, | |
| "eval_samples_per_second": 11.724, | |
| "eval_steps_per_second": 5.863, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.001266490765171504, | |
| "grad_norm": 58.67621994018555, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 3.2996, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0016886543535620053, | |
| "grad_norm": 22.526634216308594, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.414, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0016886543535620053, | |
| "eval_loss": 0.269288033246994, | |
| "eval_runtime": 425.2416, | |
| "eval_samples_per_second": 11.727, | |
| "eval_steps_per_second": 5.865, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0021108179419525065, | |
| "grad_norm": 40.498443603515625, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.8166, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.002532981530343008, | |
| "grad_norm": 13.02916431427002, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2817, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.002532981530343008, | |
| "eval_loss": 0.07939723134040833, | |
| "eval_runtime": 425.447, | |
| "eval_samples_per_second": 11.722, | |
| "eval_steps_per_second": 5.862, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0029551451187335093, | |
| "grad_norm": 25.84929656982422, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.1741, | |
| "step": 35 | |
| } | |
| ], | |
| "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": 6835420016934912.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |