| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.3401360544217687, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.006802721088435374, | |
| "eval_loss": 7.88132381439209, | |
| "eval_runtime": 37.6145, | |
| "eval_samples_per_second": 26.32, | |
| "eval_steps_per_second": 3.297, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06802721088435375, | |
| "grad_norm": 5.561442852020264, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 6.6527, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06802721088435375, | |
| "eval_loss": 3.9917244911193848, | |
| "eval_runtime": 36.513, | |
| "eval_samples_per_second": 27.114, | |
| "eval_steps_per_second": 3.396, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1360544217687075, | |
| "grad_norm": 2.8865180015563965, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 3.2544, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1360544217687075, | |
| "eval_loss": 2.308483123779297, | |
| "eval_runtime": 36.6555, | |
| "eval_samples_per_second": 27.008, | |
| "eval_steps_per_second": 3.383, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.20408163265306123, | |
| "grad_norm": 2.2314932346343994, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 1.9778, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.20408163265306123, | |
| "eval_loss": 1.6901850700378418, | |
| "eval_runtime": 36.1306, | |
| "eval_samples_per_second": 27.401, | |
| "eval_steps_per_second": 3.432, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.272108843537415, | |
| "grad_norm": 2.5240602493286133, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 1.447, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.272108843537415, | |
| "eval_loss": 1.3286805152893066, | |
| "eval_runtime": 36.3788, | |
| "eval_samples_per_second": 27.214, | |
| "eval_steps_per_second": 3.409, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3401360544217687, | |
| "grad_norm": 2.9974277019500732, | |
| "learning_rate": 0.0, | |
| "loss": 1.2756, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3401360544217687, | |
| "eval_loss": 1.2556947469711304, | |
| "eval_runtime": 36.4335, | |
| "eval_samples_per_second": 27.173, | |
| "eval_steps_per_second": 3.403, | |
| "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": 5.02662300696576e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |