| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.10075566750629723, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0020151133501259445, | |
| "eval_loss": 0.1221700981259346, | |
| "eval_runtime": 14.5639, | |
| "eval_samples_per_second": 14.351, | |
| "eval_steps_per_second": 7.21, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.020151133501259445, | |
| "grad_norm": 0.5562965869903564, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 0.4241, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02619647355163728, | |
| "eval_loss": 0.0726490169763565, | |
| "eval_runtime": 14.5687, | |
| "eval_samples_per_second": 14.346, | |
| "eval_steps_per_second": 7.207, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.04030226700251889, | |
| "grad_norm": 0.18682315945625305, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.2885, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05239294710327456, | |
| "eval_loss": 0.0567830465734005, | |
| "eval_runtime": 14.57, | |
| "eval_samples_per_second": 14.345, | |
| "eval_steps_per_second": 7.207, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.060453400503778336, | |
| "grad_norm": 0.22602833807468414, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 0.1853, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07858942065491184, | |
| "eval_loss": 0.05225418880581856, | |
| "eval_runtime": 14.5852, | |
| "eval_samples_per_second": 14.33, | |
| "eval_steps_per_second": 7.199, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.08060453400503778, | |
| "grad_norm": 0.12900914251804352, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 0.1836, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10075566750629723, | |
| "grad_norm": 0.1916407346725464, | |
| "learning_rate": 0.0, | |
| "loss": 0.1707, | |
| "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": 1.62874924204032e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |