| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.22857142857142856, | |
| "eval_steps": 3, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.009142857142857144, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.009142857142857144, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.7976, | |
| "eval_samples_per_second": 19.385, | |
| "eval_steps_per_second": 9.797, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.018285714285714287, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.027428571428571427, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.027428571428571427, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.005, | |
| "eval_samples_per_second": 23.221, | |
| "eval_steps_per_second": 11.735, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.036571428571428574, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.045714285714285714, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.054857142857142854, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.054857142857142854, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0154, | |
| "eval_samples_per_second": 23.161, | |
| "eval_steps_per_second": 11.705, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.064, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.07314285714285715, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.08228571428571428, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.08228571428571428, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.03, | |
| "eval_samples_per_second": 23.077, | |
| "eval_steps_per_second": 11.663, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.09142857142857143, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10057142857142858, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019781476007338058, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.10971428571428571, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001913545457642601, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.10971428571428571, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0214, | |
| "eval_samples_per_second": 23.126, | |
| "eval_steps_per_second": 11.687, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.11885714285714286, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.128, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016691306063588583, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.13714285714285715, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.13714285714285715, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.004, | |
| "eval_samples_per_second": 23.227, | |
| "eval_steps_per_second": 11.738, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1462857142857143, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.15542857142857142, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00011045284632676536, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.16457142857142856, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.954715367323468e-05, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.16457142857142856, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.002, | |
| "eval_samples_per_second": 23.239, | |
| "eval_steps_per_second": 11.744, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1737142857142857, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.18285714285714286, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.308693936411421e-05, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0268, | |
| "eval_samples_per_second": 23.095, | |
| "eval_steps_per_second": 11.672, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.20114285714285715, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.2102857142857143, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.645454235739903e-06, | |
| "loss": 0.0, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.21942857142857142, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.1852399266194314e-06, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.21942857142857142, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0035, | |
| "eval_samples_per_second": 23.23, | |
| "eval_steps_per_second": 11.74, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.22857142857142856, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6531344184115200.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |