| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.005759373380176237, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.759373380176237e-05, | |
| "eval_loss": 2.6716082096099854, | |
| "eval_runtime": 1331.768, | |
| "eval_samples_per_second": 21.958, | |
| "eval_steps_per_second": 2.745, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001727812014052871, | |
| "grad_norm": 1.0326002836227417, | |
| "learning_rate": 3e-05, | |
| "loss": 10.8192, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0003455624028105742, | |
| "grad_norm": 1.0763981342315674, | |
| "learning_rate": 6e-05, | |
| "loss": 10.7332, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0005183436042158614, | |
| "grad_norm": 1.2686634063720703, | |
| "learning_rate": 9e-05, | |
| "loss": 10.432, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0005183436042158614, | |
| "eval_loss": 2.629826784133911, | |
| "eval_runtime": 1338.5029, | |
| "eval_samples_per_second": 21.848, | |
| "eval_steps_per_second": 2.731, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0006911248056211485, | |
| "grad_norm": 1.1208940744400024, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 10.5925, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0008639060070264355, | |
| "grad_norm": 1.1096599102020264, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 9.8268, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0010366872084317227, | |
| "grad_norm": 1.2350603342056274, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 10.5695, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0010366872084317227, | |
| "eval_loss": 2.5420780181884766, | |
| "eval_runtime": 1337.1556, | |
| "eval_samples_per_second": 21.87, | |
| "eval_steps_per_second": 2.734, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0012094684098370098, | |
| "grad_norm": 1.1920151710510254, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 10.2468, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.001382249611242297, | |
| "grad_norm": 0.8807956576347351, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 9.8125, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.001555030812647584, | |
| "grad_norm": 0.8759644627571106, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 9.77, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.001555030812647584, | |
| "eval_loss": 2.4928576946258545, | |
| "eval_runtime": 1338.7702, | |
| "eval_samples_per_second": 21.843, | |
| "eval_steps_per_second": 2.731, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.001727812014052871, | |
| "grad_norm": 0.9653000235557556, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 10.2316, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0019005932154581582, | |
| "grad_norm": 0.8884105682373047, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 10.0839, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0020733744168634455, | |
| "grad_norm": 1.2079200744628906, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 10.2626, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0020733744168634455, | |
| "eval_loss": 2.4689135551452637, | |
| "eval_runtime": 1339.7306, | |
| "eval_samples_per_second": 21.828, | |
| "eval_steps_per_second": 2.729, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0022461556182687323, | |
| "grad_norm": 0.9944831132888794, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 9.843, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0024189368196740196, | |
| "grad_norm": 0.935939371585846, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 9.7281, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0025917180210793065, | |
| "grad_norm": 1.00493323802948, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 10.2203, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0025917180210793065, | |
| "eval_loss": 2.4538426399230957, | |
| "eval_runtime": 1338.817, | |
| "eval_samples_per_second": 21.842, | |
| "eval_steps_per_second": 2.731, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.002764499222484594, | |
| "grad_norm": 0.9063613414764404, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 9.8641, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0029372804238898807, | |
| "grad_norm": 1.0362025499343872, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 9.7865, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.003110061625295168, | |
| "grad_norm": 0.9377679824829102, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 10.0832, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.003110061625295168, | |
| "eval_loss": 2.4470882415771484, | |
| "eval_runtime": 1338.2458, | |
| "eval_samples_per_second": 21.852, | |
| "eval_steps_per_second": 2.732, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.003282842826700455, | |
| "grad_norm": 1.549514651298523, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 9.8436, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.003455624028105742, | |
| "grad_norm": 0.9654721021652222, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 10.4088, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.003628405229511029, | |
| "grad_norm": 0.9682468771934509, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 9.6252, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.003628405229511029, | |
| "eval_loss": 2.440478563308716, | |
| "eval_runtime": 1339.1783, | |
| "eval_samples_per_second": 21.837, | |
| "eval_steps_per_second": 2.73, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0038011864309163163, | |
| "grad_norm": 1.087673306465149, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 9.8558, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.003973967632321603, | |
| "grad_norm": 0.9650527834892273, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 9.4286, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.004146748833726891, | |
| "grad_norm": 1.0395511388778687, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 9.8884, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.004146748833726891, | |
| "eval_loss": 2.4370479583740234, | |
| "eval_runtime": 1337.1445, | |
| "eval_samples_per_second": 21.87, | |
| "eval_steps_per_second": 2.734, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.004319530035132178, | |
| "grad_norm": 0.8846902251243591, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 9.9641, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.004492311236537465, | |
| "grad_norm": 1.0415371656417847, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 9.3965, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.0046650924379427515, | |
| "grad_norm": 1.0671367645263672, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 9.7664, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.0046650924379427515, | |
| "eval_loss": 2.4352002143859863, | |
| "eval_runtime": 1335.8618, | |
| "eval_samples_per_second": 21.891, | |
| "eval_steps_per_second": 2.737, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.004837873639348039, | |
| "grad_norm": 1.150208830833435, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 9.6133, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.005010654840753326, | |
| "grad_norm": 0.9796428084373474, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 10.1937, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.005183436042158613, | |
| "grad_norm": 0.966742992401123, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 9.7158, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.005183436042158613, | |
| "eval_loss": 2.434671401977539, | |
| "eval_runtime": 1339.1776, | |
| "eval_samples_per_second": 21.837, | |
| "eval_steps_per_second": 2.73, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0053562172435639, | |
| "grad_norm": 0.9962571859359741, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 9.2092, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.005528998444969188, | |
| "grad_norm": 1.0119867324829102, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 9.5268, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.0057017796463744745, | |
| "grad_norm": 1.0774887800216675, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 9.4073, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.0057017796463744745, | |
| "eval_loss": 2.4344372749328613, | |
| "eval_runtime": 1338.1241, | |
| "eval_samples_per_second": 21.854, | |
| "eval_steps_per_second": 2.732, | |
| "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": 7.534154269222502e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |