| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 200, | |
| "global_step": 9160, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2183406113537118, | |
| "eval_loss": 2.382188320159912, | |
| "eval_runtime": 74.685, | |
| "eval_samples_per_second": 13.617, | |
| "eval_steps_per_second": 1.714, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4366812227074236, | |
| "eval_loss": 0.27150478959083557, | |
| "eval_runtime": 75.1491, | |
| "eval_samples_per_second": 13.533, | |
| "eval_steps_per_second": 1.703, | |
| "eval_wer": 0.5093029350104822, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5458515283842795, | |
| "grad_norm": 0.9615474939346313, | |
| "learning_rate": 0.0002868211920529801, | |
| "loss": 2.7769, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6550218340611353, | |
| "eval_loss": 0.24891583621501923, | |
| "eval_runtime": 75.3374, | |
| "eval_samples_per_second": 13.499, | |
| "eval_steps_per_second": 1.699, | |
| "eval_wer": 0.4820492662473795, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8733624454148472, | |
| "eval_loss": 0.2296326756477356, | |
| "eval_runtime": 74.8314, | |
| "eval_samples_per_second": 13.591, | |
| "eval_steps_per_second": 1.711, | |
| "eval_wer": 0.46947064989517817, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.091703056768559, | |
| "grad_norm": 0.819220244884491, | |
| "learning_rate": 0.00027026490066225163, | |
| "loss": 0.6809, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.091703056768559, | |
| "eval_loss": 0.22090712189674377, | |
| "eval_runtime": 75.0858, | |
| "eval_samples_per_second": 13.544, | |
| "eval_steps_per_second": 1.705, | |
| "eval_wer": 0.4638364779874214, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.3100436681222707, | |
| "eval_loss": 0.21630799770355225, | |
| "eval_runtime": 75.4504, | |
| "eval_samples_per_second": 13.479, | |
| "eval_steps_per_second": 1.696, | |
| "eval_wer": 0.44693396226415094, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.5283842794759825, | |
| "eval_loss": 0.20916949212551117, | |
| "eval_runtime": 75.3429, | |
| "eval_samples_per_second": 13.498, | |
| "eval_steps_per_second": 1.699, | |
| "eval_wer": 0.4399895178197065, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.6375545851528384, | |
| "grad_norm": 0.9643642902374268, | |
| "learning_rate": 0.00025370860927152316, | |
| "loss": 0.6113, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.7467248908296944, | |
| "eval_loss": 0.20470178127288818, | |
| "eval_runtime": 74.8687, | |
| "eval_samples_per_second": 13.584, | |
| "eval_steps_per_second": 1.71, | |
| "eval_wer": 0.4346174004192872, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.965065502183406, | |
| "eval_loss": 0.20738434791564941, | |
| "eval_runtime": 75.3186, | |
| "eval_samples_per_second": 13.503, | |
| "eval_steps_per_second": 1.699, | |
| "eval_wer": 0.4466719077568134, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.183406113537118, | |
| "grad_norm": 0.6018229126930237, | |
| "learning_rate": 0.00023715231788079468, | |
| "loss": 0.5974, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.183406113537118, | |
| "eval_loss": 0.20408853888511658, | |
| "eval_runtime": 75.0651, | |
| "eval_samples_per_second": 13.548, | |
| "eval_steps_per_second": 1.705, | |
| "eval_wer": 0.43042452830188677, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.4017467248908297, | |
| "eval_loss": 0.20539002120494843, | |
| "eval_runtime": 74.9069, | |
| "eval_samples_per_second": 13.577, | |
| "eval_steps_per_second": 1.709, | |
| "eval_wer": 0.4317348008385744, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.6200873362445414, | |
| "eval_loss": 0.19870109856128693, | |
| "eval_runtime": 75.2079, | |
| "eval_samples_per_second": 13.523, | |
| "eval_steps_per_second": 1.702, | |
| "eval_wer": 0.4240041928721174, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.7292576419213974, | |
| "grad_norm": 0.7469688653945923, | |
| "learning_rate": 0.0002205960264900662, | |
| "loss": 0.5636, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.8384279475982535, | |
| "eval_loss": 0.20032186806201935, | |
| "eval_runtime": 75.3206, | |
| "eval_samples_per_second": 13.502, | |
| "eval_steps_per_second": 1.699, | |
| "eval_wer": 0.42518343815513626, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.056768558951965, | |
| "eval_loss": 0.19965218007564545, | |
| "eval_runtime": 74.7464, | |
| "eval_samples_per_second": 13.606, | |
| "eval_steps_per_second": 1.712, | |
| "eval_wer": 0.42872117400419285, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.2751091703056767, | |
| "grad_norm": 0.8557049036026001, | |
| "learning_rate": 0.00020403973509933773, | |
| "loss": 0.5398, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.2751091703056767, | |
| "eval_loss": 0.20972934365272522, | |
| "eval_runtime": 74.9379, | |
| "eval_samples_per_second": 13.571, | |
| "eval_steps_per_second": 1.708, | |
| "eval_wer": 0.4399895178197065, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.493449781659389, | |
| "eval_loss": 0.19680900871753693, | |
| "eval_runtime": 74.7045, | |
| "eval_samples_per_second": 13.614, | |
| "eval_steps_per_second": 1.713, | |
| "eval_wer": 0.4165356394129979, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.7117903930131004, | |
| "eval_loss": 0.20125041902065277, | |
| "eval_runtime": 74.9279, | |
| "eval_samples_per_second": 13.573, | |
| "eval_steps_per_second": 1.708, | |
| "eval_wer": 0.42177672955974843, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 3.8209606986899565, | |
| "grad_norm": 0.7604547739028931, | |
| "learning_rate": 0.00018748344370860925, | |
| "loss": 0.5334, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.930131004366812, | |
| "eval_loss": 0.20029041171073914, | |
| "eval_runtime": 74.9149, | |
| "eval_samples_per_second": 13.575, | |
| "eval_steps_per_second": 1.709, | |
| "eval_wer": 0.4229559748427673, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.148471615720524, | |
| "eval_loss": 0.19763976335525513, | |
| "eval_runtime": 75.2231, | |
| "eval_samples_per_second": 13.52, | |
| "eval_steps_per_second": 1.702, | |
| "eval_wer": 0.4226939203354298, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 4.366812227074236, | |
| "grad_norm": 1.0875004529953003, | |
| "learning_rate": 0.00017092715231788077, | |
| "loss": 0.5123, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.366812227074236, | |
| "eval_loss": 0.19775792956352234, | |
| "eval_runtime": 75.5854, | |
| "eval_samples_per_second": 13.455, | |
| "eval_steps_per_second": 1.693, | |
| "eval_wer": 0.419811320754717, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.585152838427947, | |
| "eval_loss": 0.201936736702919, | |
| "eval_runtime": 74.8299, | |
| "eval_samples_per_second": 13.591, | |
| "eval_steps_per_second": 1.711, | |
| "eval_wer": 0.429769392033543, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 4.8034934497816595, | |
| "eval_loss": 0.19385558366775513, | |
| "eval_runtime": 75.489, | |
| "eval_samples_per_second": 13.472, | |
| "eval_steps_per_second": 1.696, | |
| "eval_wer": 0.41457023060796644, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 4.9126637554585155, | |
| "grad_norm": 1.0764209032058716, | |
| "learning_rate": 0.0001543708609271523, | |
| "loss": 0.5119, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.021834061135372, | |
| "eval_loss": 0.1989189237356186, | |
| "eval_runtime": 75.1857, | |
| "eval_samples_per_second": 13.527, | |
| "eval_steps_per_second": 1.702, | |
| "eval_wer": 0.4161425576519916, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 5.240174672489083, | |
| "eval_loss": 0.1901673972606659, | |
| "eval_runtime": 75.3755, | |
| "eval_samples_per_second": 13.492, | |
| "eval_steps_per_second": 1.698, | |
| "eval_wer": 0.407625786163522, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 5.458515283842795, | |
| "grad_norm": 0.5494298338890076, | |
| "learning_rate": 0.00013781456953642382, | |
| "loss": 0.4929, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.458515283842795, | |
| "eval_loss": 0.1928589940071106, | |
| "eval_runtime": 75.5607, | |
| "eval_samples_per_second": 13.459, | |
| "eval_steps_per_second": 1.694, | |
| "eval_wer": 0.4115566037735849, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.676855895196507, | |
| "eval_loss": 0.19432277977466583, | |
| "eval_runtime": 75.57, | |
| "eval_samples_per_second": 13.458, | |
| "eval_steps_per_second": 1.694, | |
| "eval_wer": 0.4144392033542977, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 5.895196506550218, | |
| "eval_loss": 0.19218797981739044, | |
| "eval_runtime": 75.4593, | |
| "eval_samples_per_second": 13.477, | |
| "eval_steps_per_second": 1.696, | |
| "eval_wer": 0.4106394129979036, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 6.004366812227074, | |
| "grad_norm": 0.40776267647743225, | |
| "learning_rate": 0.00012125827814569536, | |
| "loss": 0.4878, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 6.11353711790393, | |
| "eval_loss": 0.19328303635120392, | |
| "eval_runtime": 76.1865, | |
| "eval_samples_per_second": 13.349, | |
| "eval_steps_per_second": 1.68, | |
| "eval_wer": 0.41365303983228513, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 6.331877729257642, | |
| "eval_loss": 0.19196276366710663, | |
| "eval_runtime": 75.3609, | |
| "eval_samples_per_second": 13.495, | |
| "eval_steps_per_second": 1.698, | |
| "eval_wer": 0.40579140461215935, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 6.550218340611353, | |
| "grad_norm": 0.5017375946044922, | |
| "learning_rate": 0.00010470198675496688, | |
| "loss": 0.4755, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 6.550218340611353, | |
| "eval_loss": 0.19270355999469757, | |
| "eval_runtime": 75.2917, | |
| "eval_samples_per_second": 13.507, | |
| "eval_steps_per_second": 1.7, | |
| "eval_wer": 0.41705974842767296, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 6.7685589519650655, | |
| "eval_loss": 0.19197432696819305, | |
| "eval_runtime": 75.4413, | |
| "eval_samples_per_second": 13.481, | |
| "eval_steps_per_second": 1.697, | |
| "eval_wer": 0.41273584905660377, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 6.986899563318778, | |
| "eval_loss": 0.19249393045902252, | |
| "eval_runtime": 75.5242, | |
| "eval_samples_per_second": 13.466, | |
| "eval_steps_per_second": 1.695, | |
| "eval_wer": 0.40605345911949686, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 7.096069868995633, | |
| "grad_norm": 0.8388169407844543, | |
| "learning_rate": 8.81456953642384e-05, | |
| "loss": 0.475, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 7.205240174672489, | |
| "eval_loss": 0.18842807412147522, | |
| "eval_runtime": 75.7212, | |
| "eval_samples_per_second": 13.431, | |
| "eval_steps_per_second": 1.69, | |
| "eval_wer": 0.40579140461215935, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 7.423580786026201, | |
| "eval_loss": 0.19032706320285797, | |
| "eval_runtime": 75.2911, | |
| "eval_samples_per_second": 13.508, | |
| "eval_steps_per_second": 1.7, | |
| "eval_wer": 0.40697064989517817, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 7.641921397379913, | |
| "grad_norm": 1.298710584640503, | |
| "learning_rate": 7.158940397350993e-05, | |
| "loss": 0.4715, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 7.641921397379913, | |
| "eval_loss": 0.18821676075458527, | |
| "eval_runtime": 75.1357, | |
| "eval_samples_per_second": 13.536, | |
| "eval_steps_per_second": 1.704, | |
| "eval_wer": 0.3996331236897275, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 7.860262008733624, | |
| "eval_loss": 0.1881488859653473, | |
| "eval_runtime": 75.7296, | |
| "eval_samples_per_second": 13.429, | |
| "eval_steps_per_second": 1.69, | |
| "eval_wer": 0.4033018867924528, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 8.078602620087336, | |
| "eval_loss": 0.1884731650352478, | |
| "eval_runtime": 75.2978, | |
| "eval_samples_per_second": 13.506, | |
| "eval_steps_per_second": 1.7, | |
| "eval_wer": 0.40068134171907754, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 8.187772925764191, | |
| "grad_norm": 0.7714540362358093, | |
| "learning_rate": 5.5033112582781456e-05, | |
| "loss": 0.4575, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 8.296943231441048, | |
| "eval_loss": 0.1885315328836441, | |
| "eval_runtime": 76.1765, | |
| "eval_samples_per_second": 13.351, | |
| "eval_steps_per_second": 1.68, | |
| "eval_wer": 0.4015985324947589, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 8.51528384279476, | |
| "eval_loss": 0.18875299394130707, | |
| "eval_runtime": 75.8312, | |
| "eval_samples_per_second": 13.411, | |
| "eval_steps_per_second": 1.688, | |
| "eval_wer": 0.40500524109014674, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 8.733624454148472, | |
| "grad_norm": 0.7762609124183655, | |
| "learning_rate": 3.850993377483443e-05, | |
| "loss": 0.4611, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 8.733624454148472, | |
| "eval_loss": 0.18837310373783112, | |
| "eval_runtime": 76.1442, | |
| "eval_samples_per_second": 13.356, | |
| "eval_steps_per_second": 1.681, | |
| "eval_wer": 0.40461215932914046, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 8.951965065502183, | |
| "eval_loss": 0.1881391853094101, | |
| "eval_runtime": 76.0237, | |
| "eval_samples_per_second": 13.377, | |
| "eval_steps_per_second": 1.684, | |
| "eval_wer": 0.39740566037735847, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 9.170305676855895, | |
| "eval_loss": 0.1864539086818695, | |
| "eval_runtime": 75.5164, | |
| "eval_samples_per_second": 13.467, | |
| "eval_steps_per_second": 1.695, | |
| "eval_wer": 0.3955712788259958, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 9.279475982532752, | |
| "grad_norm": 0.7670572400093079, | |
| "learning_rate": 2.1953642384105956e-05, | |
| "loss": 0.4559, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 9.388646288209607, | |
| "eval_loss": 0.18750794231891632, | |
| "eval_runtime": 75.8042, | |
| "eval_samples_per_second": 13.416, | |
| "eval_steps_per_second": 1.689, | |
| "eval_wer": 0.39740566037735847, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 9.606986899563319, | |
| "eval_loss": 0.1871640682220459, | |
| "eval_runtime": 75.5158, | |
| "eval_samples_per_second": 13.467, | |
| "eval_steps_per_second": 1.695, | |
| "eval_wer": 0.3996331236897275, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 9.825327510917031, | |
| "grad_norm": 0.47464126348495483, | |
| "learning_rate": 5.3973509933774825e-06, | |
| "loss": 0.4536, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 9.825327510917031, | |
| "eval_loss": 0.18756870925426483, | |
| "eval_runtime": 75.5452, | |
| "eval_samples_per_second": 13.462, | |
| "eval_steps_per_second": 1.694, | |
| "eval_wer": 0.3953092243186583, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 9160, | |
| "total_flos": 1.4294624233263563e+19, | |
| "train_loss": 0.6386700151268572, | |
| "train_runtime": 13601.4767, | |
| "train_samples_per_second": 5.387, | |
| "train_steps_per_second": 0.673 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 9160, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4294624233263563e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |