| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.3872216844143272, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.003872216844143272, | |
| "eval_loss": 0.18226714432239532, | |
| "eval_runtime": 31.1866, | |
| "eval_samples_per_second": 13.948, | |
| "eval_steps_per_second": 1.764, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011616650532429816, | |
| "grad_norm": 4.684105396270752, | |
| "learning_rate": 3e-05, | |
| "loss": 0.7421, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.023233301064859633, | |
| "grad_norm": 3.108184576034546, | |
| "learning_rate": 6e-05, | |
| "loss": 0.4496, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.03484995159728945, | |
| "grad_norm": 1.3367609977722168, | |
| "learning_rate": 9e-05, | |
| "loss": 0.154, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.03484995159728945, | |
| "eval_loss": 0.005886069033294916, | |
| "eval_runtime": 31.6952, | |
| "eval_samples_per_second": 13.724, | |
| "eval_steps_per_second": 1.735, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.046466602129719266, | |
| "grad_norm": 0.09060771763324738, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.0114, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.05808325266214908, | |
| "grad_norm": 0.8179565072059631, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.0038, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0696999031945789, | |
| "grad_norm": 1.4341135025024414, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.0018, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0696999031945789, | |
| "eval_loss": 0.00017564959125593305, | |
| "eval_runtime": 31.7814, | |
| "eval_samples_per_second": 13.687, | |
| "eval_steps_per_second": 1.731, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.08131655372700872, | |
| "grad_norm": 0.004023493733257055, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.0001, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.09293320425943853, | |
| "grad_norm": 0.23858696222305298, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0015, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.10454985479186835, | |
| "grad_norm": 0.026785193011164665, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.0033, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.10454985479186835, | |
| "eval_loss": 0.0001601200783625245, | |
| "eval_runtime": 31.8153, | |
| "eval_samples_per_second": 13.673, | |
| "eval_steps_per_second": 1.729, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.11616650532429816, | |
| "grad_norm": 0.3270968198776245, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0015, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12778315585672798, | |
| "grad_norm": 0.01174607127904892, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.0003, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.1393998063891578, | |
| "grad_norm": 0.9263725280761719, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0008, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.1393998063891578, | |
| "eval_loss": 0.0002172403474105522, | |
| "eval_runtime": 31.8495, | |
| "eval_samples_per_second": 13.658, | |
| "eval_steps_per_second": 1.727, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.1510164569215876, | |
| "grad_norm": 0.01262608077377081, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.0014, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.16263310745401743, | |
| "grad_norm": 0.01573655568063259, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.0002, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.17424975798644723, | |
| "grad_norm": 0.061099544167518616, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.0002, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.17424975798644723, | |
| "eval_loss": 0.00012387760216370225, | |
| "eval_runtime": 31.8026, | |
| "eval_samples_per_second": 13.678, | |
| "eval_steps_per_second": 1.729, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.18586640851887706, | |
| "grad_norm": 0.05485091730952263, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.0015, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.19748305905130686, | |
| "grad_norm": 0.002600950188934803, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.0001, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.2090997095837367, | |
| "grad_norm": 0.0022585629485547543, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.0, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.2090997095837367, | |
| "eval_loss": 0.00028623713296838105, | |
| "eval_runtime": 31.8325, | |
| "eval_samples_per_second": 13.665, | |
| "eval_steps_per_second": 1.728, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.2207163601161665, | |
| "grad_norm": 0.007769421208649874, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.0001, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.23233301064859632, | |
| "grad_norm": 0.0027711803559213877, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0048, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.24394966118102615, | |
| "grad_norm": 0.003505075117573142, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.0004, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.24394966118102615, | |
| "eval_loss": 2.9244753022794612e-05, | |
| "eval_runtime": 31.8226, | |
| "eval_samples_per_second": 13.67, | |
| "eval_steps_per_second": 1.728, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.25556631171345595, | |
| "grad_norm": 0.0020100402180105448, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.0004, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.2671829622458858, | |
| "grad_norm": 0.004158449359238148, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.0001, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.2787996127783156, | |
| "grad_norm": 0.08573237806558609, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.0001, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.2787996127783156, | |
| "eval_loss": 2.5135599571513012e-05, | |
| "eval_runtime": 31.8171, | |
| "eval_samples_per_second": 13.672, | |
| "eval_steps_per_second": 1.729, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.2904162633107454, | |
| "grad_norm": 0.006858558859676123, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.0001, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.3020329138431752, | |
| "grad_norm": 0.004765031859278679, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.0001, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.31364956437560504, | |
| "grad_norm": 0.011414007283747196, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.0001, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.31364956437560504, | |
| "eval_loss": 2.312105061719194e-05, | |
| "eval_runtime": 31.8058, | |
| "eval_samples_per_second": 13.677, | |
| "eval_steps_per_second": 1.729, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.32526621490803487, | |
| "grad_norm": 0.007097288966178894, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.0001, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.33688286544046464, | |
| "grad_norm": 0.002511974424123764, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.0001, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.34849951597289447, | |
| "grad_norm": 0.0035338769666850567, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.0001, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.34849951597289447, | |
| "eval_loss": 2.103859333146829e-05, | |
| "eval_runtime": 31.8073, | |
| "eval_samples_per_second": 13.676, | |
| "eval_steps_per_second": 1.729, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3601161665053243, | |
| "grad_norm": 0.3512181043624878, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.0002, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.3717328170377541, | |
| "grad_norm": 0.005195324309170246, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.0001, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.38334946757018395, | |
| "grad_norm": 0.07296066731214523, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.0001, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.38334946757018395, | |
| "eval_loss": 1.9248587705078535e-05, | |
| "eval_runtime": 31.8075, | |
| "eval_samples_per_second": 13.676, | |
| "eval_steps_per_second": 1.729, | |
| "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": 1.4279806347431117e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |