| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.006492347145801886, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.492347145801887e-05, | |
| "eval_loss": 1.2574450969696045, | |
| "eval_runtime": 2164.1177, | |
| "eval_samples_per_second": 11.987, | |
| "eval_steps_per_second": 1.499, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00019477041437405659, | |
| "grad_norm": 0.21717190742492676, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1457, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00038954082874811317, | |
| "grad_norm": 0.1491047739982605, | |
| "learning_rate": 6e-05, | |
| "loss": 1.2315, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0005843112431221698, | |
| "grad_norm": 0.18038997054100037, | |
| "learning_rate": 9e-05, | |
| "loss": 1.1694, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0005843112431221698, | |
| "eval_loss": 1.246630072593689, | |
| "eval_runtime": 2177.4551, | |
| "eval_samples_per_second": 11.914, | |
| "eval_steps_per_second": 1.489, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0007790816574962263, | |
| "grad_norm": 0.2281239777803421, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.3245, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0009738520718702829, | |
| "grad_norm": 0.24129489064216614, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.2447, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0011686224862443395, | |
| "grad_norm": 0.20848846435546875, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 1.3351, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0011686224862443395, | |
| "eval_loss": 1.1892844438552856, | |
| "eval_runtime": 2174.5469, | |
| "eval_samples_per_second": 11.93, | |
| "eval_steps_per_second": 1.491, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.001363392900618396, | |
| "grad_norm": 0.22571416199207306, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 1.1832, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0015581633149924527, | |
| "grad_norm": 0.24647393822669983, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.3137, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0017529337293665092, | |
| "grad_norm": 0.2367832213640213, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 1.24, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0017529337293665092, | |
| "eval_loss": 1.1527782678604126, | |
| "eval_runtime": 2174.8129, | |
| "eval_samples_per_second": 11.928, | |
| "eval_steps_per_second": 1.491, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0019477041437405659, | |
| "grad_norm": 0.2434782236814499, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 1.0845, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0021424745581146223, | |
| "grad_norm": 0.1957402527332306, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 1.0744, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.002337244972488679, | |
| "grad_norm": 0.20614418387413025, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 1.0835, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.002337244972488679, | |
| "eval_loss": 1.136653184890747, | |
| "eval_runtime": 2176.2744, | |
| "eval_samples_per_second": 11.92, | |
| "eval_steps_per_second": 1.49, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0025320153868627357, | |
| "grad_norm": 0.19403059780597687, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 1.2458, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.002726785801236792, | |
| "grad_norm": 0.2524452805519104, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 1.0825, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0029215562156108487, | |
| "grad_norm": 0.22213758528232574, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 1.1252, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0029215562156108487, | |
| "eval_loss": 1.1249250173568726, | |
| "eval_runtime": 2174.4018, | |
| "eval_samples_per_second": 11.931, | |
| "eval_steps_per_second": 1.491, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0031163266299849054, | |
| "grad_norm": 0.3361480236053467, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 1.1268, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.003311097044358962, | |
| "grad_norm": 0.26797690987586975, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 1.0806, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.0035058674587330183, | |
| "grad_norm": 0.24251677095890045, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 1.1435, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.0035058674587330183, | |
| "eval_loss": 1.1167911291122437, | |
| "eval_runtime": 2174.2458, | |
| "eval_samples_per_second": 11.931, | |
| "eval_steps_per_second": 1.492, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.003700637873107075, | |
| "grad_norm": 0.25566375255584717, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 1.1187, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.0038954082874811317, | |
| "grad_norm": 0.22987186908721924, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 1.1929, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.004090178701855188, | |
| "grad_norm": 0.2270929515361786, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 1.061, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.004090178701855188, | |
| "eval_loss": 1.1118545532226562, | |
| "eval_runtime": 2176.6142, | |
| "eval_samples_per_second": 11.919, | |
| "eval_steps_per_second": 1.49, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.004284949116229245, | |
| "grad_norm": 0.24837514758110046, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 1.0437, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.004479719530603301, | |
| "grad_norm": 0.2640354037284851, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.9759, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.004674489944977358, | |
| "grad_norm": 0.2336353361606598, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 1.009, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.004674489944977358, | |
| "eval_loss": 1.107908844947815, | |
| "eval_runtime": 2174.4566, | |
| "eval_samples_per_second": 11.93, | |
| "eval_steps_per_second": 1.491, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.004869260359351414, | |
| "grad_norm": 0.2683839201927185, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 1.0181, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.0050640307737254715, | |
| "grad_norm": 0.2877131998538971, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 1.0598, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.005258801188099528, | |
| "grad_norm": 0.278920441865921, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 1.1018, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.005258801188099528, | |
| "eval_loss": 1.1061383485794067, | |
| "eval_runtime": 2176.8672, | |
| "eval_samples_per_second": 11.917, | |
| "eval_steps_per_second": 1.49, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.005453571602473584, | |
| "grad_norm": 0.2761317491531372, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 1.024, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.005648342016847641, | |
| "grad_norm": 0.29402637481689453, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 1.2208, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.005843112431221697, | |
| "grad_norm": 0.1755310297012329, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.951, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.005843112431221697, | |
| "eval_loss": 1.1051793098449707, | |
| "eval_runtime": 2177.0043, | |
| "eval_samples_per_second": 11.916, | |
| "eval_steps_per_second": 1.49, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.006037882845595754, | |
| "grad_norm": 0.2584977447986603, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 1.0409, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.006232653259969811, | |
| "grad_norm": 0.21517957746982574, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 1.102, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.006427423674343867, | |
| "grad_norm": 0.2704929709434509, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 1.256, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.006427423674343867, | |
| "eval_loss": 1.1049902439117432, | |
| "eval_runtime": 2176.592, | |
| "eval_samples_per_second": 11.919, | |
| "eval_steps_per_second": 1.49, | |
| "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": 1.6401962573286605e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |