| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.05422993492407809, | |
| "eval_steps": 4, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0010845986984815619, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0010845986984815619, | |
| "eval_loss": NaN, | |
| "eval_runtime": 51.7259, | |
| "eval_samples_per_second": 7.52, | |
| "eval_steps_per_second": 3.77, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0021691973969631237, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0032537960954446853, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004338394793926247, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.004338394793926247, | |
| "eval_loss": NaN, | |
| "eval_runtime": 54.4666, | |
| "eval_samples_per_second": 7.142, | |
| "eval_steps_per_second": 3.58, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.005422993492407809, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.006507592190889371, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.007592190889370932, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.008676789587852495, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.008676789587852495, | |
| "eval_loss": NaN, | |
| "eval_runtime": 80.2469, | |
| "eval_samples_per_second": 4.848, | |
| "eval_steps_per_second": 2.43, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.009761388286334056, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.010845986984815618, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01193058568329718, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.013015184381778741, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.013015184381778741, | |
| "eval_loss": NaN, | |
| "eval_runtime": 50.8068, | |
| "eval_samples_per_second": 7.656, | |
| "eval_steps_per_second": 3.838, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.014099783080260303, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.015184381778741865, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.016268980477223426, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.01735357917570499, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.01735357917570499, | |
| "eval_loss": NaN, | |
| "eval_runtime": 50.8293, | |
| "eval_samples_per_second": 7.653, | |
| "eval_steps_per_second": 3.836, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.01843817787418655, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.019522776572668113, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.020607375271149676, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.021691973969631236, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.021691973969631236, | |
| "eval_loss": NaN, | |
| "eval_runtime": 50.8357, | |
| "eval_samples_per_second": 7.652, | |
| "eval_steps_per_second": 3.836, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0227765726681128, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.02386117136659436, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.024945770065075923, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.0, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.026030368763557483, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.026030368763557483, | |
| "eval_loss": NaN, | |
| "eval_runtime": 50.847, | |
| "eval_samples_per_second": 7.65, | |
| "eval_steps_per_second": 3.835, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.027114967462039046, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.028199566160520606, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.02928416485900217, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03036876355748373, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.03036876355748373, | |
| "eval_loss": NaN, | |
| "eval_runtime": 107.1539, | |
| "eval_samples_per_second": 3.63, | |
| "eval_steps_per_second": 1.82, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.031453362255965296, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.0, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.03253796095444685, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.033622559652928416, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.0, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.03470715835140998, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.0, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.03470715835140998, | |
| "eval_loss": NaN, | |
| "eval_runtime": 50.8923, | |
| "eval_samples_per_second": 7.644, | |
| "eval_steps_per_second": 3.832, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.03579175704989154, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0368763557483731, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.03796095444685466, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.039045553145336226, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.039045553145336226, | |
| "eval_loss": NaN, | |
| "eval_runtime": 50.8731, | |
| "eval_samples_per_second": 7.646, | |
| "eval_steps_per_second": 3.833, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.04013015184381779, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.0, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.04121475054229935, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.0, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.04229934924078091, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.04338394793926247, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04338394793926247, | |
| "eval_loss": NaN, | |
| "eval_runtime": 50.7973, | |
| "eval_samples_per_second": 7.658, | |
| "eval_steps_per_second": 3.839, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.044468546637744036, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.0, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0455531453362256, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.046637744034707156, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.0, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.04772234273318872, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.0, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.04772234273318872, | |
| "eval_loss": NaN, | |
| "eval_runtime": 50.8611, | |
| "eval_samples_per_second": 7.648, | |
| "eval_steps_per_second": 3.834, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.04880694143167028, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.049891540130151846, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.0509761388286334, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 0.0, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.052060737527114966, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.052060737527114966, | |
| "eval_loss": NaN, | |
| "eval_runtime": 50.9078, | |
| "eval_samples_per_second": 7.641, | |
| "eval_steps_per_second": 3.83, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.05314533622559653, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 0.0, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.05422993492407809, | |
| "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": 1.39742146461696e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |