| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.043215211754537596, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.000432152117545376, | |
| "eval_loss": 2.312277317047119, | |
| "eval_runtime": 180.5436, | |
| "eval_samples_per_second": 86.345, | |
| "eval_steps_per_second": 2.703, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0012964563526361278, | |
| "grad_norm": 2.2848575115203857, | |
| "learning_rate": 3e-05, | |
| "loss": 2.362, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0025929127052722557, | |
| "grad_norm": 1.7597225904464722, | |
| "learning_rate": 6e-05, | |
| "loss": 2.1744, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0038893690579083835, | |
| "grad_norm": 1.3628700971603394, | |
| "learning_rate": 9e-05, | |
| "loss": 1.9805, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0038893690579083835, | |
| "eval_loss": 1.8443766832351685, | |
| "eval_runtime": 132.1231, | |
| "eval_samples_per_second": 117.988, | |
| "eval_steps_per_second": 3.694, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.005185825410544511, | |
| "grad_norm": 1.3412178754806519, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.829, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.00648228176318064, | |
| "grad_norm": 1.5588266849517822, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.7529, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.007778738115816767, | |
| "grad_norm": 1.1395800113677979, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 1.7135, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.007778738115816767, | |
| "eval_loss": 1.6603822708129883, | |
| "eval_runtime": 132.0674, | |
| "eval_samples_per_second": 118.038, | |
| "eval_steps_per_second": 3.695, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.009075194468452896, | |
| "grad_norm": 1.0279678106307983, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 1.6145, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.010371650821089023, | |
| "grad_norm": 0.9357244372367859, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.6543, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.011668107173725151, | |
| "grad_norm": 0.962315022945404, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 1.6006, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.011668107173725151, | |
| "eval_loss": 1.5944404602050781, | |
| "eval_runtime": 132.0865, | |
| "eval_samples_per_second": 118.021, | |
| "eval_steps_per_second": 3.695, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.01296456352636128, | |
| "grad_norm": 0.888576865196228, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 1.578, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.014261019878997408, | |
| "grad_norm": 0.9320745468139648, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 1.5553, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.015557476231633534, | |
| "grad_norm": 0.9765816330909729, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 1.5462, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.015557476231633534, | |
| "eval_loss": 1.5527358055114746, | |
| "eval_runtime": 132.0946, | |
| "eval_samples_per_second": 118.014, | |
| "eval_steps_per_second": 3.694, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.016853932584269662, | |
| "grad_norm": 0.9101667404174805, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 1.5658, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.018150388936905792, | |
| "grad_norm": 0.9854418635368347, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 1.4873, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.01944684528954192, | |
| "grad_norm": 1.0120340585708618, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 1.5698, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.01944684528954192, | |
| "eval_loss": 1.5253695249557495, | |
| "eval_runtime": 132.0927, | |
| "eval_samples_per_second": 118.016, | |
| "eval_steps_per_second": 3.694, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.020743301642178046, | |
| "grad_norm": 0.9587317705154419, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 1.5692, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.022039757994814176, | |
| "grad_norm": 0.9989524483680725, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 1.5135, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.023336214347450302, | |
| "grad_norm": 1.1253865957260132, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 1.5211, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.023336214347450302, | |
| "eval_loss": 1.503990888595581, | |
| "eval_runtime": 132.1403, | |
| "eval_samples_per_second": 117.973, | |
| "eval_steps_per_second": 3.693, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.024632670700086432, | |
| "grad_norm": 0.997772753238678, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 1.5019, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.02592912705272256, | |
| "grad_norm": 0.9458330869674683, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 1.4898, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.027225583405358685, | |
| "grad_norm": 0.9065344929695129, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 1.5003, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.027225583405358685, | |
| "eval_loss": 1.4908527135849, | |
| "eval_runtime": 132.1066, | |
| "eval_samples_per_second": 118.003, | |
| "eval_steps_per_second": 3.694, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.028522039757994815, | |
| "grad_norm": 0.8747905492782593, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 1.4359, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.029818496110630942, | |
| "grad_norm": 0.9813586473464966, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 1.5078, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.03111495246326707, | |
| "grad_norm": 0.9811728596687317, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 1.5199, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.03111495246326707, | |
| "eval_loss": 1.4821799993515015, | |
| "eval_runtime": 132.105, | |
| "eval_samples_per_second": 118.005, | |
| "eval_steps_per_second": 3.694, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.032411408815903195, | |
| "grad_norm": 1.0040117502212524, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 1.5108, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.033707865168539325, | |
| "grad_norm": 0.8985682129859924, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 1.458, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.035004321521175455, | |
| "grad_norm": 0.9384897351264954, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 1.5195, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.035004321521175455, | |
| "eval_loss": 1.4772651195526123, | |
| "eval_runtime": 132.0692, | |
| "eval_samples_per_second": 118.037, | |
| "eval_steps_per_second": 3.695, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.036300777873811585, | |
| "grad_norm": 0.978691577911377, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 1.4837, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.03759723422644771, | |
| "grad_norm": 0.992375910282135, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 1.4403, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.03889369057908384, | |
| "grad_norm": 1.0013164281845093, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 1.4748, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.03889369057908384, | |
| "eval_loss": 1.47517728805542, | |
| "eval_runtime": 132.0325, | |
| "eval_samples_per_second": 118.069, | |
| "eval_steps_per_second": 3.696, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.04019014693171997, | |
| "grad_norm": 0.984551191329956, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 1.4686, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.04148660328435609, | |
| "grad_norm": 1.0442090034484863, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 1.5048, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.04278305963699222, | |
| "grad_norm": 0.8811266422271729, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 1.4768, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.04278305963699222, | |
| "eval_loss": 1.4746631383895874, | |
| "eval_runtime": 132.1003, | |
| "eval_samples_per_second": 118.009, | |
| "eval_steps_per_second": 3.694, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.170710885236736e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |