| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.13178036605657237, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0013311148086522463, | |
| "eval_loss": 5.295779705047607, | |
| "eval_runtime": 93.7316, | |
| "eval_samples_per_second": 13.496, | |
| "eval_steps_per_second": 1.696, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.003993344425956739, | |
| "grad_norm": 6.330746650695801, | |
| "learning_rate": 3e-05, | |
| "loss": 5.2865, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.007986688851913478, | |
| "grad_norm": 5.513250350952148, | |
| "learning_rate": 6e-05, | |
| "loss": 4.7382, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.011980033277870216, | |
| "grad_norm": 4.898468017578125, | |
| "learning_rate": 9e-05, | |
| "loss": 3.1585, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.011980033277870216, | |
| "eval_loss": 2.123068332672119, | |
| "eval_runtime": 94.2958, | |
| "eval_samples_per_second": 13.415, | |
| "eval_steps_per_second": 1.686, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.015973377703826955, | |
| "grad_norm": 5.28930139541626, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 2.0198, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.019966722129783693, | |
| "grad_norm": 5.305483341217041, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.5995, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02396006655574043, | |
| "grad_norm": 3.3500654697418213, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 1.2395, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02396006655574043, | |
| "eval_loss": 0.9006592631340027, | |
| "eval_runtime": 94.3026, | |
| "eval_samples_per_second": 13.414, | |
| "eval_steps_per_second": 1.686, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.027953410981697173, | |
| "grad_norm": 2.782661199569702, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.8408, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.03194675540765391, | |
| "grad_norm": 4.695487022399902, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.5291, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03594009983361065, | |
| "grad_norm": 4.321674346923828, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.5393, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03594009983361065, | |
| "eval_loss": 0.36986345052719116, | |
| "eval_runtime": 94.3472, | |
| "eval_samples_per_second": 13.408, | |
| "eval_steps_per_second": 1.685, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03993344425956739, | |
| "grad_norm": 2.1721081733703613, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.3295, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.043926788685524125, | |
| "grad_norm": 1.9686952829360962, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.2444, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.04792013311148086, | |
| "grad_norm": 3.1062581539154053, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.3405, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.04792013311148086, | |
| "eval_loss": 0.22597768902778625, | |
| "eval_runtime": 94.2858, | |
| "eval_samples_per_second": 13.417, | |
| "eval_steps_per_second": 1.686, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0519134775374376, | |
| "grad_norm": 2.1065447330474854, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.2113, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.055906821963394346, | |
| "grad_norm": 2.7513046264648438, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.1744, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.059900166389351084, | |
| "grad_norm": 2.247255802154541, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.1895, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.059900166389351084, | |
| "eval_loss": 0.15109963715076447, | |
| "eval_runtime": 94.3125, | |
| "eval_samples_per_second": 13.413, | |
| "eval_steps_per_second": 1.686, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06389351081530782, | |
| "grad_norm": 2.3737082481384277, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.2152, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.06788685524126456, | |
| "grad_norm": 2.481839656829834, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.1393, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.0718801996672213, | |
| "grad_norm": 4.425018787384033, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.159, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.0718801996672213, | |
| "eval_loss": 0.12050312012434006, | |
| "eval_runtime": 94.2539, | |
| "eval_samples_per_second": 13.421, | |
| "eval_steps_per_second": 1.687, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.07587354409317804, | |
| "grad_norm": 1.1071043014526367, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.1544, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.07986688851913477, | |
| "grad_norm": 1.475027322769165, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0855, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.08386023294509151, | |
| "grad_norm": 1.2657231092453003, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.1186, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.08386023294509151, | |
| "eval_loss": 0.10414409637451172, | |
| "eval_runtime": 94.3116, | |
| "eval_samples_per_second": 13.413, | |
| "eval_steps_per_second": 1.686, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.08785357737104825, | |
| "grad_norm": 2.2030110359191895, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.1065, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.09184692179700499, | |
| "grad_norm": 1.347221851348877, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.0879, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.09584026622296173, | |
| "grad_norm": 1.6131019592285156, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.1761, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.09584026622296173, | |
| "eval_loss": 0.08786367624998093, | |
| "eval_runtime": 94.295, | |
| "eval_samples_per_second": 13.415, | |
| "eval_steps_per_second": 1.686, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.09983361064891846, | |
| "grad_norm": 1.8289293050765991, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.0966, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1038269550748752, | |
| "grad_norm": 1.5840426683425903, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.0604, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.10782029950083195, | |
| "grad_norm": 1.362238883972168, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.0767, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.10782029950083195, | |
| "eval_loss": 0.08052682131528854, | |
| "eval_runtime": 94.2717, | |
| "eval_samples_per_second": 13.419, | |
| "eval_steps_per_second": 1.687, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.11181364392678869, | |
| "grad_norm": 2.0179669857025146, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.0702, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.11580698835274543, | |
| "grad_norm": 1.466431975364685, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.0989, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.11980033277870217, | |
| "grad_norm": 1.6172847747802734, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.1468, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.11980033277870217, | |
| "eval_loss": 0.07370365411043167, | |
| "eval_runtime": 94.2945, | |
| "eval_samples_per_second": 13.415, | |
| "eval_steps_per_second": 1.686, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1237936772046589, | |
| "grad_norm": 0.7145574688911438, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.0618, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.12778702163061564, | |
| "grad_norm": 0.8748223185539246, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.0521, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.13178036605657237, | |
| "grad_norm": 1.464680790901184, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.0948, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.13178036605657237, | |
| "eval_loss": 0.07193463295698166, | |
| "eval_runtime": 94.3059, | |
| "eval_samples_per_second": 13.414, | |
| "eval_steps_per_second": 1.686, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.33230992818176e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |