| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0161290322580645, | |
| "eval_steps": 200, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06720430107526881, | |
| "eval_loss": 3.080493211746216, | |
| "eval_runtime": 198.5178, | |
| "eval_samples_per_second": 35.372, | |
| "eval_steps_per_second": 0.554, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.13440860215053763, | |
| "eval_loss": 2.92264461517334, | |
| "eval_runtime": 201.3326, | |
| "eval_samples_per_second": 34.878, | |
| "eval_steps_per_second": 0.546, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.16801075268817203, | |
| "grad_norm": 5.131893634796143, | |
| "learning_rate": 0.00024599999999999996, | |
| "loss": 4.4646, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.20161290322580644, | |
| "eval_loss": 1.3392444849014282, | |
| "eval_runtime": 198.1589, | |
| "eval_samples_per_second": 35.436, | |
| "eval_steps_per_second": 0.555, | |
| "eval_wer": 0.8646141989662366, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.26881720430107525, | |
| "eval_loss": 1.0115001201629639, | |
| "eval_runtime": 200.6717, | |
| "eval_samples_per_second": 34.992, | |
| "eval_steps_per_second": 0.548, | |
| "eval_wer": 0.7168179221089837, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.33602150537634407, | |
| "grad_norm": 6.313941478729248, | |
| "learning_rate": 0.0002782777777777778, | |
| "loss": 1.1042, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.33602150537634407, | |
| "eval_loss": 0.9008844494819641, | |
| "eval_runtime": 199.6418, | |
| "eval_samples_per_second": 35.173, | |
| "eval_steps_per_second": 0.551, | |
| "eval_wer": 0.6766765436691957, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4032258064516129, | |
| "eval_loss": 0.8225907683372498, | |
| "eval_runtime": 199.7483, | |
| "eval_samples_per_second": 35.154, | |
| "eval_steps_per_second": 0.551, | |
| "eval_wer": 0.5947727182131041, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.47043010752688175, | |
| "eval_loss": 0.8068214058876038, | |
| "eval_runtime": 197.9234, | |
| "eval_samples_per_second": 35.478, | |
| "eval_steps_per_second": 0.556, | |
| "eval_wer": 0.5977358189718174, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.5040322580645161, | |
| "grad_norm": 8.026273727416992, | |
| "learning_rate": 0.0002505555555555555, | |
| "loss": 0.8421, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.5376344086021505, | |
| "eval_loss": 0.7186344861984253, | |
| "eval_runtime": 197.8905, | |
| "eval_samples_per_second": 35.484, | |
| "eval_steps_per_second": 0.556, | |
| "eval_wer": 0.5375171075884612, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.6048387096774194, | |
| "eval_loss": 0.7200700640678406, | |
| "eval_runtime": 197.0513, | |
| "eval_samples_per_second": 35.635, | |
| "eval_steps_per_second": 0.558, | |
| "eval_wer": 0.5560796715343016, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.6720430107526881, | |
| "grad_norm": 4.496196746826172, | |
| "learning_rate": 0.00022283333333333332, | |
| "loss": 0.7637, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6720430107526881, | |
| "eval_loss": 0.6747044920921326, | |
| "eval_runtime": 198.0259, | |
| "eval_samples_per_second": 35.46, | |
| "eval_steps_per_second": 0.555, | |
| "eval_wer": 0.5086966342895867, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.739247311827957, | |
| "eval_loss": 0.6709386706352234, | |
| "eval_runtime": 199.0402, | |
| "eval_samples_per_second": 35.279, | |
| "eval_steps_per_second": 0.553, | |
| "eval_wer": 0.5050558737161004, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.8064516129032258, | |
| "eval_loss": 0.6484177708625793, | |
| "eval_runtime": 193.7002, | |
| "eval_samples_per_second": 36.252, | |
| "eval_steps_per_second": 0.568, | |
| "eval_wer": 0.48504497800927465, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.8400537634408602, | |
| "grad_norm": 4.925416469573975, | |
| "learning_rate": 0.00019516666666666665, | |
| "loss": 0.7134, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.8736559139784946, | |
| "eval_loss": 0.6387702226638794, | |
| "eval_runtime": 191.5475, | |
| "eval_samples_per_second": 36.659, | |
| "eval_steps_per_second": 0.574, | |
| "eval_wer": 0.4915159648679892, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.9408602150537635, | |
| "eval_loss": 0.627938985824585, | |
| "eval_runtime": 192.2841, | |
| "eval_samples_per_second": 36.519, | |
| "eval_steps_per_second": 0.572, | |
| "eval_wer": 0.4705217980573752, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.0080645161290323, | |
| "grad_norm": 2.991081714630127, | |
| "learning_rate": 0.00016738888888888888, | |
| "loss": 0.6968, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0080645161290323, | |
| "eval_loss": 0.6451985836029053, | |
| "eval_runtime": 192.2109, | |
| "eval_samples_per_second": 36.533, | |
| "eval_steps_per_second": 0.572, | |
| "eval_wer": 0.4833441847486679, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.075268817204301, | |
| "eval_loss": 0.6439228057861328, | |
| "eval_runtime": 191.7283, | |
| "eval_samples_per_second": 36.625, | |
| "eval_steps_per_second": 0.574, | |
| "eval_wer": 0.4896955845812461, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.14247311827957, | |
| "eval_loss": 0.6749800443649292, | |
| "eval_runtime": 191.0444, | |
| "eval_samples_per_second": 36.756, | |
| "eval_steps_per_second": 0.576, | |
| "eval_wer": 0.5010164897221595, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.1760752688172043, | |
| "grad_norm": 17.283771514892578, | |
| "learning_rate": 0.00013977777777777776, | |
| "loss": 0.667, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.2096774193548387, | |
| "eval_loss": 0.7904173135757446, | |
| "eval_runtime": 191.7132, | |
| "eval_samples_per_second": 36.628, | |
| "eval_steps_per_second": 0.574, | |
| "eval_wer": 0.5591756467664997, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.2768817204301075, | |
| "eval_loss": 0.9479840993881226, | |
| "eval_runtime": 192.8834, | |
| "eval_samples_per_second": 36.405, | |
| "eval_steps_per_second": 0.57, | |
| "eval_wer": 0.5800502265509773, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.3440860215053765, | |
| "grad_norm": 26.34053611755371, | |
| "learning_rate": 0.0001121111111111111, | |
| "loss": 0.9326, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.3440860215053765, | |
| "eval_loss": 1.1457170248031616, | |
| "eval_runtime": 189.9612, | |
| "eval_samples_per_second": 36.965, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.6581804169600978, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.4112903225806452, | |
| "eval_loss": 1.118348479270935, | |
| "eval_runtime": 190.0092, | |
| "eval_samples_per_second": 36.956, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.6742183659097251, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.478494623655914, | |
| "eval_loss": 1.3061094284057617, | |
| "eval_runtime": 190.4579, | |
| "eval_samples_per_second": 36.869, | |
| "eval_steps_per_second": 0.578, | |
| "eval_wer": 0.78073054385522, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.5120967741935485, | |
| "grad_norm": 15.129014015197754, | |
| "learning_rate": 8.433333333333331e-05, | |
| "loss": 1.2308, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.5456989247311828, | |
| "eval_loss": 1.2298626899719238, | |
| "eval_runtime": 190.6991, | |
| "eval_samples_per_second": 36.822, | |
| "eval_steps_per_second": 0.577, | |
| "eval_wer": 0.7672304973491543, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.6129032258064515, | |
| "eval_loss": 1.178423523902893, | |
| "eval_runtime": 190.1804, | |
| "eval_samples_per_second": 36.923, | |
| "eval_steps_per_second": 0.578, | |
| "eval_wer": 0.7516974713987696, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 1.6801075268817205, | |
| "grad_norm": 3.6261146068573, | |
| "learning_rate": 5.6555555555555555e-05, | |
| "loss": 1.2835, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.6801075268817205, | |
| "eval_loss": 1.339849829673767, | |
| "eval_runtime": 189.6315, | |
| "eval_samples_per_second": 37.03, | |
| "eval_steps_per_second": 0.58, | |
| "eval_wer": 0.7625267409877888, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.7473118279569892, | |
| "eval_loss": 1.5754368305206299, | |
| "eval_runtime": 190.0111, | |
| "eval_samples_per_second": 36.956, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.895573951288218, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 1.814516129032258, | |
| "eval_loss": 1.728023886680603, | |
| "eval_runtime": 190.3066, | |
| "eval_samples_per_second": 36.898, | |
| "eval_steps_per_second": 0.578, | |
| "eval_wer": 0.9705550166757464, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 1.8481182795698925, | |
| "grad_norm": 4.430452346801758, | |
| "learning_rate": 2.8777777777777776e-05, | |
| "loss": 1.6466, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.881720430107527, | |
| "eval_loss": 1.7603241205215454, | |
| "eval_runtime": 190.0315, | |
| "eval_samples_per_second": 36.952, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.9727208705935503, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.9489247311827957, | |
| "eval_loss": 1.8200371265411377, | |
| "eval_runtime": 190.0189, | |
| "eval_samples_per_second": 36.954, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.9943661223242403, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 2.0161290322580645, | |
| "grad_norm": 8.303234100341797, | |
| "learning_rate": 1.0555555555555555e-06, | |
| "loss": 1.8511, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.0161290322580645, | |
| "eval_loss": 1.8412970304489136, | |
| "eval_runtime": 189.9189, | |
| "eval_samples_per_second": 36.974, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.9956417172696953, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.0161290322580645, | |
| "step": 6000, | |
| "total_flos": 1.3379687773166643e+19, | |
| "train_loss": 1.3496905517578126, | |
| "train_runtime": 9965.7858, | |
| "train_samples_per_second": 9.633, | |
| "train_steps_per_second": 0.602 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 1.3379687773166643e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |