| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.22857142857142856, | |
| "eval_steps": 4, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004571428571428572, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004571428571428572, | |
| "eval_loss": NaN, | |
| "eval_runtime": 5.3028, | |
| "eval_samples_per_second": 17.538, | |
| "eval_steps_per_second": 8.863, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.009142857142857144, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.013714285714285714, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.018285714285714287, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.018285714285714287, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0599, | |
| "eval_samples_per_second": 22.907, | |
| "eval_steps_per_second": 11.577, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.022857142857142857, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.027428571428571427, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.032, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.036571428571428574, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.036571428571428574, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0771, | |
| "eval_samples_per_second": 22.811, | |
| "eval_steps_per_second": 11.528, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.04114285714285714, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.045714285714285714, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05028571428571429, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.054857142857142854, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.054857142857142854, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0449, | |
| "eval_samples_per_second": 22.992, | |
| "eval_steps_per_second": 11.62, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.05942857142857143, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.064, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.06857142857142857, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.07314285714285715, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.07314285714285715, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0629, | |
| "eval_samples_per_second": 22.89, | |
| "eval_steps_per_second": 11.568, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.07771428571428571, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.08228571428571428, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.08685714285714285, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.09142857142857143, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09142857142857143, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0605, | |
| "eval_samples_per_second": 22.904, | |
| "eval_steps_per_second": 11.575, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.10057142857142858, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.10514285714285715, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.0, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.10971428571428571, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.10971428571428571, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0909, | |
| "eval_samples_per_second": 22.733, | |
| "eval_steps_per_second": 11.489, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.11428571428571428, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.11885714285714286, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.12342857142857143, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.128, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.128, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0737, | |
| "eval_samples_per_second": 22.829, | |
| "eval_steps_per_second": 11.537, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.13257142857142856, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.0, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.13714285714285715, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1417142857142857, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.0, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.1462857142857143, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.0, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.1462857142857143, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0607, | |
| "eval_samples_per_second": 22.903, | |
| "eval_steps_per_second": 11.574, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.15085714285714286, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.15542857142857142, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.16457142857142856, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.16457142857142856, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0438, | |
| "eval_samples_per_second": 22.998, | |
| "eval_steps_per_second": 11.623, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.16914285714285715, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.0, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.1737142857142857, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.0, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.1782857142857143, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.18285714285714286, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.18285714285714286, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0452, | |
| "eval_samples_per_second": 22.99, | |
| "eval_steps_per_second": 11.619, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.18742857142857142, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.0, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.19657142857142856, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.0, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.20114285714285715, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.0, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.20114285714285715, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0451, | |
| "eval_samples_per_second": 22.991, | |
| "eval_steps_per_second": 11.619, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.2057142857142857, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2102857142857143, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.21485714285714286, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 0.0, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.21942857142857142, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.21942857142857142, | |
| "eval_loss": NaN, | |
| "eval_runtime": 4.0555, | |
| "eval_samples_per_second": 22.932, | |
| "eval_steps_per_second": 11.589, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.224, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 0.0, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.22857142857142856, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 4, | |
| "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 | |
| } | |