| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.3333333333333333, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.013468013468013467, | |
| "eval_loss": 0.4762314558029175, | |
| "eval_runtime": 11.1179, | |
| "eval_samples_per_second": 11.243, | |
| "eval_steps_per_second": 1.439, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04040404040404041, | |
| "grad_norm": 4.391507148742676, | |
| "learning_rate": 3e-05, | |
| "loss": 0.479, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08080808080808081, | |
| "grad_norm": 1.9744468927383423, | |
| "learning_rate": 6e-05, | |
| "loss": 0.3377, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.12121212121212122, | |
| "grad_norm": 0.8217880129814148, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0948, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.12121212121212122, | |
| "eval_loss": 0.006129190791398287, | |
| "eval_runtime": 11.2698, | |
| "eval_samples_per_second": 11.092, | |
| "eval_steps_per_second": 1.42, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.16161616161616163, | |
| "grad_norm": 0.07933423668146133, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.003, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.20202020202020202, | |
| "grad_norm": 0.016316216439008713, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.0004, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.24242424242424243, | |
| "grad_norm": 0.0206525269895792, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.0003, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.24242424242424243, | |
| "eval_loss": 0.0002653657866176218, | |
| "eval_runtime": 11.2826, | |
| "eval_samples_per_second": 11.079, | |
| "eval_steps_per_second": 1.418, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2828282828282828, | |
| "grad_norm": 0.008192160166800022, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.0002, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.32323232323232326, | |
| "grad_norm": 0.002964617684483528, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0001, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.0018415789818391204, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "eval_loss": 3.535100768203847e-05, | |
| "eval_runtime": 11.2827, | |
| "eval_samples_per_second": 11.079, | |
| "eval_steps_per_second": 1.418, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.40404040404040403, | |
| "grad_norm": 0.0016876906156539917, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 0.001268128864467144, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.48484848484848486, | |
| "grad_norm": 0.0012457451084628701, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.48484848484848486, | |
| "eval_loss": 2.0379013221827336e-05, | |
| "eval_runtime": 11.2858, | |
| "eval_samples_per_second": 11.076, | |
| "eval_steps_per_second": 1.418, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5252525252525253, | |
| "grad_norm": 0.0010147533612325788, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.5656565656565656, | |
| "grad_norm": 0.0007608148152939975, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.6060606060606061, | |
| "grad_norm": 0.0006441783043555915, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6060606060606061, | |
| "eval_loss": 1.3838871382176876e-05, | |
| "eval_runtime": 11.276, | |
| "eval_samples_per_second": 11.085, | |
| "eval_steps_per_second": 1.419, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6464646464646465, | |
| "grad_norm": 0.0005675142747350037, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.6868686868686869, | |
| "grad_norm": 0.00048168693319894373, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.0, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.00044461351353675127, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.0, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "eval_loss": 1.0228068276774138e-05, | |
| "eval_runtime": 11.2817, | |
| "eval_samples_per_second": 11.08, | |
| "eval_steps_per_second": 1.418, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.7676767676767676, | |
| "grad_norm": 0.00037802866427227855, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.0, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.8080808080808081, | |
| "grad_norm": 0.00036502134753391147, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8484848484848485, | |
| "grad_norm": 0.00033542574965395033, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.0, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.8484848484848485, | |
| "eval_loss": 8.655947567604017e-06, | |
| "eval_runtime": 11.275, | |
| "eval_samples_per_second": 11.086, | |
| "eval_steps_per_second": 1.419, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 0.00031755914096720517, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.0, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.9292929292929293, | |
| "grad_norm": 0.00030167968361638486, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.0, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.9696969696969697, | |
| "grad_norm": 0.0002713320136535913, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.0, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.9696969696969697, | |
| "eval_loss": 7.80674508860102e-06, | |
| "eval_runtime": 11.3027, | |
| "eval_samples_per_second": 11.059, | |
| "eval_steps_per_second": 1.416, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.0101010101010102, | |
| "grad_norm": 0.0005017324001528323, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.0505050505050506, | |
| "grad_norm": 0.0002725861268118024, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.0, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 0.0002557328261900693, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.0, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "eval_loss": 7.262304734467762e-06, | |
| "eval_runtime": 11.2928, | |
| "eval_samples_per_second": 11.069, | |
| "eval_steps_per_second": 1.417, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.1313131313131313, | |
| "grad_norm": 0.0002512714418116957, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.0, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.1717171717171717, | |
| "grad_norm": 0.00025135595933534205, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.0, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.2121212121212122, | |
| "grad_norm": 0.00025588576681911945, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.2121212121212122, | |
| "eval_loss": 7.283484137587948e-06, | |
| "eval_runtime": 11.2861, | |
| "eval_samples_per_second": 11.076, | |
| "eval_steps_per_second": 1.418, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.2525252525252526, | |
| "grad_norm": 0.00024979611043818295, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.0, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.2929292929292928, | |
| "grad_norm": 0.0002572893863543868, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.0, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.0002596003469079733, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.0, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "eval_loss": 7.183366960816784e-06, | |
| "eval_runtime": 11.2842, | |
| "eval_samples_per_second": 11.077, | |
| "eval_steps_per_second": 1.418, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.677619125855191e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |