| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 1682, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0011890606420927466, | |
| "grad_norm": 115.31934356689453, | |
| "learning_rate": 4.9970273483947685e-05, | |
| "loss": 9.3546, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.034482758620689655, | |
| "grad_norm": 18.599306106567383, | |
| "learning_rate": 4.913793103448276e-05, | |
| "loss": 2.4091, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.06896551724137931, | |
| "grad_norm": 12.393567085266113, | |
| "learning_rate": 4.827586206896552e-05, | |
| "loss": 1.1241, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.10344827586206896, | |
| "grad_norm": 50.094051361083984, | |
| "learning_rate": 4.741379310344828e-05, | |
| "loss": 1.0125, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.13793103448275862, | |
| "grad_norm": 39.95038986206055, | |
| "learning_rate": 4.655172413793104e-05, | |
| "loss": 0.9273, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.1724137931034483, | |
| "grad_norm": 13.281441688537598, | |
| "learning_rate": 4.5689655172413794e-05, | |
| "loss": 0.8734, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.20689655172413793, | |
| "grad_norm": 29.820255279541016, | |
| "learning_rate": 4.482758620689655e-05, | |
| "loss": 0.9182, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 0.2413793103448276, | |
| "grad_norm": 14.038665771484375, | |
| "learning_rate": 4.396551724137931e-05, | |
| "loss": 0.9129, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 0.27586206896551724, | |
| "grad_norm": 13.406801223754883, | |
| "learning_rate": 4.3103448275862066e-05, | |
| "loss": 0.7973, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.3103448275862069, | |
| "grad_norm": 15.321793556213379, | |
| "learning_rate": 4.224137931034483e-05, | |
| "loss": 0.8415, | |
| "step": 261 | |
| }, | |
| { | |
| "epoch": 0.3448275862068966, | |
| "grad_norm": 27.121288299560547, | |
| "learning_rate": 4.1379310344827587e-05, | |
| "loss": 0.7995, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.3793103448275862, | |
| "grad_norm": 36.715858459472656, | |
| "learning_rate": 4.0517241379310344e-05, | |
| "loss": 0.6301, | |
| "step": 319 | |
| }, | |
| { | |
| "epoch": 0.41379310344827586, | |
| "grad_norm": 16.650150299072266, | |
| "learning_rate": 3.965517241379311e-05, | |
| "loss": 0.5675, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 0.4482758620689655, | |
| "grad_norm": 230.8839569091797, | |
| "learning_rate": 3.8793103448275865e-05, | |
| "loss": 0.6154, | |
| "step": 377 | |
| }, | |
| { | |
| "epoch": 0.4827586206896552, | |
| "grad_norm": 6.414034366607666, | |
| "learning_rate": 3.793103448275862e-05, | |
| "loss": 0.6439, | |
| "step": 406 | |
| }, | |
| { | |
| "epoch": 0.5172413793103449, | |
| "grad_norm": 9.441433906555176, | |
| "learning_rate": 3.7068965517241385e-05, | |
| "loss": 0.5776, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 0.5517241379310345, | |
| "grad_norm": 23.88970947265625, | |
| "learning_rate": 3.620689655172414e-05, | |
| "loss": 0.6126, | |
| "step": 464 | |
| }, | |
| { | |
| "epoch": 0.5862068965517241, | |
| "grad_norm": 10.72873592376709, | |
| "learning_rate": 3.53448275862069e-05, | |
| "loss": 0.529, | |
| "step": 493 | |
| }, | |
| { | |
| "epoch": 0.6206896551724138, | |
| "grad_norm": 22.285120010375977, | |
| "learning_rate": 3.4482758620689657e-05, | |
| "loss": 0.5418, | |
| "step": 522 | |
| }, | |
| { | |
| "epoch": 0.6551724137931034, | |
| "grad_norm": 28.53759002685547, | |
| "learning_rate": 3.3620689655172414e-05, | |
| "loss": 0.4864, | |
| "step": 551 | |
| }, | |
| { | |
| "epoch": 0.6896551724137931, | |
| "grad_norm": 15.215431213378906, | |
| "learning_rate": 3.275862068965517e-05, | |
| "loss": 0.5474, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.7241379310344828, | |
| "grad_norm": 5.726695537567139, | |
| "learning_rate": 3.1896551724137935e-05, | |
| "loss": 0.4492, | |
| "step": 609 | |
| }, | |
| { | |
| "epoch": 0.7586206896551724, | |
| "grad_norm": 20.181598663330078, | |
| "learning_rate": 3.103448275862069e-05, | |
| "loss": 0.4925, | |
| "step": 638 | |
| }, | |
| { | |
| "epoch": 0.7931034482758621, | |
| "grad_norm": 16.79625701904297, | |
| "learning_rate": 3.017241379310345e-05, | |
| "loss": 0.4405, | |
| "step": 667 | |
| }, | |
| { | |
| "epoch": 0.8275862068965517, | |
| "grad_norm": 5.420645236968994, | |
| "learning_rate": 2.9310344827586206e-05, | |
| "loss": 0.5411, | |
| "step": 696 | |
| }, | |
| { | |
| "epoch": 0.8620689655172413, | |
| "grad_norm": 6.218174457550049, | |
| "learning_rate": 2.844827586206897e-05, | |
| "loss": 0.5516, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.896551724137931, | |
| "grad_norm": 27.71928596496582, | |
| "learning_rate": 2.7586206896551727e-05, | |
| "loss": 0.4341, | |
| "step": 754 | |
| }, | |
| { | |
| "epoch": 0.9310344827586207, | |
| "grad_norm": 31.43971824645996, | |
| "learning_rate": 2.672413793103448e-05, | |
| "loss": 0.3883, | |
| "step": 783 | |
| }, | |
| { | |
| "epoch": 0.9655172413793104, | |
| "grad_norm": 7.160225868225098, | |
| "learning_rate": 2.5862068965517244e-05, | |
| "loss": 0.4214, | |
| "step": 812 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 7.139071464538574, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.3728, | |
| "step": 841 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_cer": 0.019845431962462048, | |
| "eval_loss": 0.2980102300643921, | |
| "eval_runtime": 887.0299, | |
| "eval_samples_per_second": 1.896, | |
| "eval_steps_per_second": 0.238, | |
| "step": 841 | |
| }, | |
| { | |
| "epoch": 1.0344827586206897, | |
| "grad_norm": 14.740863800048828, | |
| "learning_rate": 2.413793103448276e-05, | |
| "loss": 0.3343, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.0689655172413792, | |
| "grad_norm": 22.295618057250977, | |
| "learning_rate": 2.327586206896552e-05, | |
| "loss": 0.3419, | |
| "step": 899 | |
| }, | |
| { | |
| "epoch": 1.103448275862069, | |
| "grad_norm": 2.4551053047180176, | |
| "learning_rate": 2.2413793103448276e-05, | |
| "loss": 0.3123, | |
| "step": 928 | |
| }, | |
| { | |
| "epoch": 1.1379310344827587, | |
| "grad_norm": 7.861540794372559, | |
| "learning_rate": 2.1551724137931033e-05, | |
| "loss": 0.349, | |
| "step": 957 | |
| }, | |
| { | |
| "epoch": 1.1724137931034484, | |
| "grad_norm": 7.037637233734131, | |
| "learning_rate": 2.0689655172413793e-05, | |
| "loss": 0.2869, | |
| "step": 986 | |
| }, | |
| { | |
| "epoch": 1.206896551724138, | |
| "grad_norm": 11.390913009643555, | |
| "learning_rate": 1.9827586206896554e-05, | |
| "loss": 0.2826, | |
| "step": 1015 | |
| }, | |
| { | |
| "epoch": 1.2413793103448276, | |
| "grad_norm": 2.8627371788024902, | |
| "learning_rate": 1.896551724137931e-05, | |
| "loss": 0.2573, | |
| "step": 1044 | |
| }, | |
| { | |
| "epoch": 1.2758620689655173, | |
| "grad_norm": 16.846853256225586, | |
| "learning_rate": 1.810344827586207e-05, | |
| "loss": 0.2752, | |
| "step": 1073 | |
| }, | |
| { | |
| "epoch": 1.3103448275862069, | |
| "grad_norm": 3.9017491340637207, | |
| "learning_rate": 1.7241379310344828e-05, | |
| "loss": 0.2555, | |
| "step": 1102 | |
| }, | |
| { | |
| "epoch": 1.3448275862068966, | |
| "grad_norm": 5.074669361114502, | |
| "learning_rate": 1.6379310344827585e-05, | |
| "loss": 0.2452, | |
| "step": 1131 | |
| }, | |
| { | |
| "epoch": 1.3793103448275863, | |
| "grad_norm": 18.69165802001953, | |
| "learning_rate": 1.5517241379310346e-05, | |
| "loss": 0.2666, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 1.4137931034482758, | |
| "grad_norm": 5.539306640625, | |
| "learning_rate": 1.4655172413793103e-05, | |
| "loss": 0.2741, | |
| "step": 1189 | |
| }, | |
| { | |
| "epoch": 1.4482758620689655, | |
| "grad_norm": 5.758066654205322, | |
| "learning_rate": 1.3793103448275863e-05, | |
| "loss": 0.2365, | |
| "step": 1218 | |
| }, | |
| { | |
| "epoch": 1.4827586206896552, | |
| "grad_norm": 4.248748779296875, | |
| "learning_rate": 1.2931034482758622e-05, | |
| "loss": 0.2521, | |
| "step": 1247 | |
| }, | |
| { | |
| "epoch": 1.5172413793103448, | |
| "grad_norm": 11.129036903381348, | |
| "learning_rate": 1.206896551724138e-05, | |
| "loss": 0.2257, | |
| "step": 1276 | |
| }, | |
| { | |
| "epoch": 1.5517241379310345, | |
| "grad_norm": 2.8656809329986572, | |
| "learning_rate": 1.1206896551724138e-05, | |
| "loss": 0.2106, | |
| "step": 1305 | |
| }, | |
| { | |
| "epoch": 1.5862068965517242, | |
| "grad_norm": 2.041245460510254, | |
| "learning_rate": 1.0344827586206897e-05, | |
| "loss": 0.227, | |
| "step": 1334 | |
| }, | |
| { | |
| "epoch": 1.6206896551724137, | |
| "grad_norm": 1.451377034187317, | |
| "learning_rate": 9.482758620689655e-06, | |
| "loss": 0.1961, | |
| "step": 1363 | |
| }, | |
| { | |
| "epoch": 1.6551724137931034, | |
| "grad_norm": 4.332240104675293, | |
| "learning_rate": 8.620689655172414e-06, | |
| "loss": 0.195, | |
| "step": 1392 | |
| }, | |
| { | |
| "epoch": 1.6896551724137931, | |
| "grad_norm": 5.428757667541504, | |
| "learning_rate": 7.758620689655173e-06, | |
| "loss": 0.1929, | |
| "step": 1421 | |
| }, | |
| { | |
| "epoch": 1.7241379310344827, | |
| "grad_norm": 2.5995798110961914, | |
| "learning_rate": 6.896551724137932e-06, | |
| "loss": 0.1991, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.7586206896551724, | |
| "grad_norm": 2.2909860610961914, | |
| "learning_rate": 6.03448275862069e-06, | |
| "loss": 0.168, | |
| "step": 1479 | |
| }, | |
| { | |
| "epoch": 1.793103448275862, | |
| "grad_norm": 2.049956798553467, | |
| "learning_rate": 5.172413793103448e-06, | |
| "loss": 0.18, | |
| "step": 1508 | |
| }, | |
| { | |
| "epoch": 1.8275862068965516, | |
| "grad_norm": 2.291414737701416, | |
| "learning_rate": 4.310344827586207e-06, | |
| "loss": 0.1642, | |
| "step": 1537 | |
| }, | |
| { | |
| "epoch": 1.8620689655172413, | |
| "grad_norm": 6.125259876251221, | |
| "learning_rate": 3.448275862068966e-06, | |
| "loss": 0.197, | |
| "step": 1566 | |
| }, | |
| { | |
| "epoch": 1.896551724137931, | |
| "grad_norm": 2.359379291534424, | |
| "learning_rate": 2.586206896551724e-06, | |
| "loss": 0.1615, | |
| "step": 1595 | |
| }, | |
| { | |
| "epoch": 1.9310344827586206, | |
| "grad_norm": 2.1470141410827637, | |
| "learning_rate": 1.724137931034483e-06, | |
| "loss": 0.1669, | |
| "step": 1624 | |
| }, | |
| { | |
| "epoch": 1.9655172413793105, | |
| "grad_norm": 13.040664672851562, | |
| "learning_rate": 8.620689655172415e-07, | |
| "loss": 0.1641, | |
| "step": 1653 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.5605217218399048, | |
| "learning_rate": 0.0, | |
| "loss": 0.1551, | |
| "step": 1682 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_cer": 0.003201766491857577, | |
| "eval_loss": 0.17416498064994812, | |
| "eval_runtime": 882.6649, | |
| "eval_samples_per_second": 1.906, | |
| "eval_steps_per_second": 0.239, | |
| "step": 1682 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 1682, | |
| "total_flos": 1.9906356553640313e+19, | |
| "train_loss": 0.47363961749354666, | |
| "train_runtime": 3260.3696, | |
| "train_samples_per_second": 4.125, | |
| "train_steps_per_second": 0.516 | |
| } | |
| ], | |
| "logging_steps": 29, | |
| "max_steps": 1682, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.9906356553640313e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |