| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.001566833273271391, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.1336665465427825e-05, | |
| "eval_loss": 11.808765411376953, | |
| "eval_runtime": 79.535, | |
| "eval_samples_per_second": 168.944, | |
| "eval_steps_per_second": 84.478, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00031336665465427825, | |
| "grad_norm": 5.586214542388916, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 11.4956, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0004073766510505617, | |
| "eval_loss": 10.812637329101562, | |
| "eval_runtime": 79.2863, | |
| "eval_samples_per_second": 169.474, | |
| "eval_steps_per_second": 84.743, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0006267333093085565, | |
| "grad_norm": 3.855236530303955, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 10.7788, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0008147533021011234, | |
| "eval_loss": 10.018383979797363, | |
| "eval_runtime": 78.6151, | |
| "eval_samples_per_second": 170.921, | |
| "eval_steps_per_second": 85.467, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0009400999639628347, | |
| "grad_norm": 3.682607650756836, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 10.0837, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0012221299531516852, | |
| "eval_loss": 9.67300033569336, | |
| "eval_runtime": 78.9736, | |
| "eval_samples_per_second": 170.145, | |
| "eval_steps_per_second": 85.079, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.001253466618617113, | |
| "grad_norm": 3.035902500152588, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 9.7752, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.001566833273271391, | |
| "grad_norm": 3.280996799468994, | |
| "learning_rate": 0.0, | |
| "loss": 9.644, | |
| "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": 3294316855296.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |