| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.84, | |
| "eval_steps": 500, | |
| "global_step": 36, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6469, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0001995184726672197, | |
| "loss": 0.4913, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 0.23202306032180786, | |
| "eval_runtime": 6.7844, | |
| "eval_samples_per_second": 14.74, | |
| "eval_steps_per_second": 1.916, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.00018314696123025454, | |
| "loss": 0.2397, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0001471396736825998, | |
| "loss": 0.2051, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_loss": 0.20802158117294312, | |
| "eval_runtime": 4.7703, | |
| "eval_samples_per_second": 20.963, | |
| "eval_steps_per_second": 2.725, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1798, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 5.286032631740023e-05, | |
| "loss": 0.1665, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_loss": 0.20379072427749634, | |
| "eval_runtime": 5.6795, | |
| "eval_samples_per_second": 17.607, | |
| "eval_steps_per_second": 2.289, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 1.6853038769745467e-05, | |
| "loss": 0.1613, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 4.815273327803182e-07, | |
| "loss": 0.1519, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "eval_loss": 0.2034250646829605, | |
| "eval_runtime": 5.6541, | |
| "eval_samples_per_second": 17.686, | |
| "eval_steps_per_second": 2.299, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "step": 36, | |
| "total_flos": 38498471247872.0, | |
| "train_loss": 0.23011421908934912, | |
| "train_runtime": 528.713, | |
| "train_samples_per_second": 2.27, | |
| "train_steps_per_second": 0.068 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 36, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "total_flos": 38498471247872.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |