| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.008128098837681867, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0001625619767536373, | |
| "eval_loss": 1.515768051147461, | |
| "eval_runtime": 88.586, | |
| "eval_samples_per_second": 29.248, | |
| "eval_steps_per_second": 14.63, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016256197675363732, | |
| "grad_norm": 2.576862096786499, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 6.0765, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.002113305697797285, | |
| "eval_loss": 1.3564525842666626, | |
| "eval_runtime": 88.5625, | |
| "eval_samples_per_second": 29.256, | |
| "eval_steps_per_second": 14.634, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0032512395350727465, | |
| "grad_norm": 2.630268096923828, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 5.5416, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00422661139559457, | |
| "eval_loss": 1.2527743577957153, | |
| "eval_runtime": 88.6178, | |
| "eval_samples_per_second": 29.238, | |
| "eval_steps_per_second": 14.625, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.00487685930260912, | |
| "grad_norm": 1.45906400680542, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 5.1671, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0063399170933918555, | |
| "eval_loss": 1.2318778038024902, | |
| "eval_runtime": 88.6117, | |
| "eval_samples_per_second": 29.24, | |
| "eval_steps_per_second": 14.626, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.006502479070145493, | |
| "grad_norm": 2.2924132347106934, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 4.9732, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.008128098837681867, | |
| "grad_norm": 1.6607002019882202, | |
| "learning_rate": 0.0, | |
| "loss": 4.8543, | |
| "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": 8835964638068736.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |