| { | |
| "best_metric": 9.4224620303757, | |
| "best_model_checkpoint": "kotoba_v2_enc_logs_epoch2_2/checkpoint-500", | |
| "epoch": 0.0017133497358014708, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.4266994716029415e-05, | |
| "grad_norm": 1.0561553239822388, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2361, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.853398943205883e-05, | |
| "grad_norm": 1.1626238822937012, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2265, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00010280098414808825, | |
| "grad_norm": 0.9845689535140991, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2279, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.00013706797886411766, | |
| "grad_norm": 1.142356276512146, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2382, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.00017133497358014707, | |
| "grad_norm": 1.0053240060806274, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2473, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0002056019682961765, | |
| "grad_norm": 1.1098105907440186, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2438, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0002398689630122059, | |
| "grad_norm": 1.191983699798584, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2293, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.0002741359577282353, | |
| "grad_norm": 1.1295104026794434, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2362, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.0003084029524442647, | |
| "grad_norm": 1.037972092628479, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2455, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.00034266994716029413, | |
| "grad_norm": 1.1975648403167725, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2459, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.00037693694187632354, | |
| "grad_norm": 1.0676342248916626, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2271, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.000411203936592353, | |
| "grad_norm": 1.0749495029449463, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2417, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.0004454709313083824, | |
| "grad_norm": 1.094260811805725, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2354, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.0004797379260244118, | |
| "grad_norm": 1.0395853519439697, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2381, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.0005140049207404412, | |
| "grad_norm": 1.2008885145187378, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2354, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.0005482719154564706, | |
| "grad_norm": 1.0647832155227661, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2321, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.0005825389101725, | |
| "grad_norm": 1.327071189880371, | |
| "learning_rate": 1e-05, | |
| "loss": 0.238, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.0006168059048885295, | |
| "grad_norm": 1.1184055805206299, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2242, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.0006510728996045589, | |
| "grad_norm": 1.2512784004211426, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2437, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.0006853398943205883, | |
| "grad_norm": 1.0614465475082397, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2382, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.0007196068890366177, | |
| "grad_norm": 1.0607149600982666, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2381, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.0007538738837526471, | |
| "grad_norm": 1.0422028303146362, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2294, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.0007881408784686765, | |
| "grad_norm": 1.0162984132766724, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2275, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.000822407873184706, | |
| "grad_norm": 1.1085543632507324, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2161, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.0008566748679007354, | |
| "grad_norm": 1.1854636669158936, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2382, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.0008909418626167648, | |
| "grad_norm": 1.40137779712677, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2579, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.0009252088573327942, | |
| "grad_norm": 1.0814112424850464, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2612, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.0009594758520488236, | |
| "grad_norm": 1.083736538887024, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2711, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.000993742846764853, | |
| "grad_norm": 1.0861411094665527, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2642, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.0010280098414808825, | |
| "grad_norm": 1.1141265630722046, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2585, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.0010622768361969119, | |
| "grad_norm": 1.326241374015808, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2858, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.0010965438309129413, | |
| "grad_norm": 1.393750786781311, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2635, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.0011308108256289707, | |
| "grad_norm": 1.0851459503173828, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2565, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.001165077820345, | |
| "grad_norm": 1.2323757410049438, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2465, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.0011993448150610295, | |
| "grad_norm": 1.376953125, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2671, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.001233611809777059, | |
| "grad_norm": 1.084592580795288, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2643, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.0012678788044930883, | |
| "grad_norm": 1.2907005548477173, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2584, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.0013021457992091177, | |
| "grad_norm": 1.0698130130767822, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2526, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.0013364127939251471, | |
| "grad_norm": 1.1399807929992676, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2759, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.0013706797886411765, | |
| "grad_norm": 1.1480791568756104, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2499, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.001404946783357206, | |
| "grad_norm": 1.3095237016677856, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2536, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.0014392137780732353, | |
| "grad_norm": 1.068246841430664, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2604, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.0014734807727892648, | |
| "grad_norm": 1.2310419082641602, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2632, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.0015077477675052942, | |
| "grad_norm": 1.161867380142212, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2584, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.0015420147622213236, | |
| "grad_norm": 1.1461217403411865, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2592, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.001576281756937353, | |
| "grad_norm": 1.3006030321121216, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2607, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.0016105487516533824, | |
| "grad_norm": 1.1223125457763672, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2433, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.001644815746369412, | |
| "grad_norm": 1.2909380197525024, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2693, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.0016790827410854414, | |
| "grad_norm": 1.2270597219467163, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2661, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.0017133497358014708, | |
| "grad_norm": 1.1439770460128784, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2517, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.0017133497358014708, | |
| "eval_cer": 13.0358087846181, | |
| "eval_loss": 0.25224336981773376, | |
| "eval_normalized_cer": 9.4224620303757, | |
| "eval_runtime": 227.2174, | |
| "eval_samples_per_second": 2.253, | |
| "eval_steps_per_second": 0.035, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 291826, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0137951535104e+20, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |