| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.136986301369863, | |
| "eval_steps": 3, | |
| "global_step": 10, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0136986301369863, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 4.7847, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0136986301369863, | |
| "eval_loss": 4.887735843658447, | |
| "eval_runtime": 1.7789, | |
| "eval_samples_per_second": 69.146, | |
| "eval_steps_per_second": 8.995, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0273972602739726, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 4.636, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0410958904109589, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 4.5614, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0410958904109589, | |
| "eval_loss": 4.887735843658447, | |
| "eval_runtime": 1.2587, | |
| "eval_samples_per_second": 97.721, | |
| "eval_steps_per_second": 12.712, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0547945205479452, | |
| "grad_norm": 10.800760269165039, | |
| "learning_rate": 2e-05, | |
| "loss": 4.8914, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0684931506849315, | |
| "grad_norm": 12.797863006591797, | |
| "learning_rate": 4e-05, | |
| "loss": 5.0571, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0821917808219178, | |
| "grad_norm": 11.154678344726562, | |
| "learning_rate": 6e-05, | |
| "loss": 4.6968, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0821917808219178, | |
| "eval_loss": 4.427567958831787, | |
| "eval_runtime": 1.2513, | |
| "eval_samples_per_second": 98.297, | |
| "eval_steps_per_second": 12.787, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0958904109589041, | |
| "grad_norm": 10.823620796203613, | |
| "learning_rate": 8e-05, | |
| "loss": 4.4948, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.1095890410958904, | |
| "grad_norm": 10.48684024810791, | |
| "learning_rate": 0.0001, | |
| "loss": 3.8815, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1232876712328767, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 3.2615, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1232876712328767, | |
| "eval_loss": 3.383413553237915, | |
| "eval_runtime": 1.2385, | |
| "eval_samples_per_second": 99.313, | |
| "eval_steps_per_second": 12.919, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.136986301369863, | |
| "grad_norm": 10.866177558898926, | |
| "learning_rate": 0.00012, | |
| "loss": 3.3794, | |
| "step": 10 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 10, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2484925806673920.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |