| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 6.789999999999999e-06, | |
| "loss": 15.499, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 1.379e-05, | |
| "loss": 5.6988, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 2.0789999999999996e-05, | |
| "loss": 3.9135, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 2.779e-05, | |
| "loss": 3.315, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 3.479e-05, | |
| "loss": 3.0713, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 4.178999999999999e-05, | |
| "loss": 2.9727, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "learning_rate": 4.878999999999999e-05, | |
| "loss": 2.8912, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 5.579e-05, | |
| "loss": 2.8514, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "learning_rate": 6.279e-05, | |
| "loss": 2.5765, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 6.979e-05, | |
| "loss": 1.785, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "eval_loss": 0.7464718222618103, | |
| "eval_runtime": 42.2657, | |
| "eval_samples_per_second": 28.013, | |
| "eval_steps_per_second": 1.751, | |
| "eval_wer": 0.6812214190883611, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 13.75, | |
| "learning_rate": 6.903e-05, | |
| "loss": 1.4773, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 6.803e-05, | |
| "loss": 1.3507, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 16.25, | |
| "learning_rate": 6.702999999999999e-05, | |
| "loss": 1.2818, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "learning_rate": 6.602999999999999e-05, | |
| "loss": 1.233, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "learning_rate": 6.502999999999999e-05, | |
| "loss": 1.161, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 6.403e-05, | |
| "loss": 1.1004, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 21.25, | |
| "learning_rate": 6.303e-05, | |
| "loss": 1.0246, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 22.5, | |
| "learning_rate": 6.202999999999999e-05, | |
| "loss": 0.9693, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 23.75, | |
| "learning_rate": 6.103e-05, | |
| "loss": 0.9507, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 6.002999999999999e-05, | |
| "loss": 0.8989, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.24948318302631378, | |
| "eval_runtime": 40.7976, | |
| "eval_samples_per_second": 29.021, | |
| "eval_steps_per_second": 1.814, | |
| "eval_wer": 0.27319663667207555, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 26.25, | |
| "learning_rate": 5.904e-05, | |
| "loss": 0.8743, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 27.5, | |
| "learning_rate": 5.804e-05, | |
| "loss": 0.8558, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 28.75, | |
| "learning_rate": 5.7039999999999996e-05, | |
| "loss": 0.8228, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 5.604e-05, | |
| "loss": 0.8154, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 31.25, | |
| "learning_rate": 5.5039999999999995e-05, | |
| "loss": 0.7932, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 32.5, | |
| "learning_rate": 5.404e-05, | |
| "loss": 0.7755, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 33.75, | |
| "learning_rate": 5.3039999999999994e-05, | |
| "loss": 0.7585, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "learning_rate": 5.2039999999999996e-05, | |
| "loss": 0.7472, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 36.25, | |
| "learning_rate": 5.103999999999999e-05, | |
| "loss": 0.734, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 37.5, | |
| "learning_rate": 5.0039999999999995e-05, | |
| "loss": 0.7118, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 37.5, | |
| "eval_loss": 0.21255508065223694, | |
| "eval_runtime": 40.7166, | |
| "eval_samples_per_second": 29.079, | |
| "eval_steps_per_second": 1.817, | |
| "eval_wer": 0.22835226434577371, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 38.75, | |
| "learning_rate": 4.904e-05, | |
| "loss": 0.7163, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 4.8039999999999994e-05, | |
| "loss": 0.7066, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 41.25, | |
| "learning_rate": 4.704e-05, | |
| "loss": 0.7067, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 42.5, | |
| "learning_rate": 4.603999999999999e-05, | |
| "loss": 0.688, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 43.75, | |
| "learning_rate": 4.5039999999999996e-05, | |
| "loss": 0.6777, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "learning_rate": 4.403999999999999e-05, | |
| "loss": 0.6621, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 46.25, | |
| "learning_rate": 4.3039999999999994e-05, | |
| "loss": 0.6494, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 47.5, | |
| "learning_rate": 4.203999999999999e-05, | |
| "loss": 0.646, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 48.75, | |
| "learning_rate": 4.104e-05, | |
| "loss": 0.6356, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 4.0039999999999996e-05, | |
| "loss": 0.6367, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 0.20487311482429504, | |
| "eval_runtime": 43.4458, | |
| "eval_samples_per_second": 27.252, | |
| "eval_steps_per_second": 1.703, | |
| "eval_wer": 0.20489747750405665, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 51.25, | |
| "learning_rate": 3.904e-05, | |
| "loss": 0.6272, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 52.5, | |
| "learning_rate": 3.804e-05, | |
| "loss": 0.6187, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 53.75, | |
| "learning_rate": 3.704e-05, | |
| "loss": 0.6116, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "learning_rate": 3.604e-05, | |
| "loss": 0.6074, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 56.25, | |
| "learning_rate": 3.504999999999999e-05, | |
| "loss": 0.5945, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 57.5, | |
| "learning_rate": 3.4049999999999994e-05, | |
| "loss": 0.6074, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 58.75, | |
| "learning_rate": 3.305e-05, | |
| "loss": 0.6011, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "learning_rate": 3.205e-05, | |
| "loss": 0.5815, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 61.25, | |
| "learning_rate": 3.1049999999999996e-05, | |
| "loss": 0.5704, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 62.5, | |
| "learning_rate": 3.0049999999999995e-05, | |
| "loss": 0.5763, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 62.5, | |
| "eval_loss": 0.21160605549812317, | |
| "eval_runtime": 41.2219, | |
| "eval_samples_per_second": 28.723, | |
| "eval_steps_per_second": 1.795, | |
| "eval_wer": 0.20548753503466588, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 63.75, | |
| "learning_rate": 2.9049999999999995e-05, | |
| "loss": 0.5742, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "learning_rate": 2.8049999999999997e-05, | |
| "loss": 0.58, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 66.25, | |
| "learning_rate": 2.705e-05, | |
| "loss": 0.5542, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 67.5, | |
| "learning_rate": 2.605e-05, | |
| "loss": 0.5472, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 68.75, | |
| "learning_rate": 2.505e-05, | |
| "loss": 0.5433, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "learning_rate": 2.405e-05, | |
| "loss": 0.5448, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 71.25, | |
| "learning_rate": 2.3049999999999998e-05, | |
| "loss": 0.5469, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 72.5, | |
| "learning_rate": 2.2049999999999997e-05, | |
| "loss": 0.5424, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 73.75, | |
| "learning_rate": 2.1049999999999997e-05, | |
| "loss": 0.5221, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "learning_rate": 2.0049999999999996e-05, | |
| "loss": 0.5196, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_loss": 0.21106497943401337, | |
| "eval_runtime": 41.193, | |
| "eval_samples_per_second": 28.743, | |
| "eval_steps_per_second": 1.796, | |
| "eval_wer": 0.19103112553473964, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 76.25, | |
| "learning_rate": 1.905e-05, | |
| "loss": 0.5215, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 77.5, | |
| "learning_rate": 1.8049999999999998e-05, | |
| "loss": 0.5141, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 78.75, | |
| "learning_rate": 1.7049999999999998e-05, | |
| "loss": 0.5087, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "learning_rate": 1.6049999999999997e-05, | |
| "loss": 0.5136, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 81.25, | |
| "learning_rate": 1.5049999999999998e-05, | |
| "loss": 0.4973, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 82.5, | |
| "learning_rate": 1.4049999999999998e-05, | |
| "loss": 0.5064, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 83.75, | |
| "learning_rate": 1.3049999999999999e-05, | |
| "loss": 0.4985, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "learning_rate": 1.205e-05, | |
| "loss": 0.4926, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 86.25, | |
| "learning_rate": 1.105e-05, | |
| "loss": 0.4973, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 87.5, | |
| "learning_rate": 1.0049999999999999e-05, | |
| "loss": 0.4949, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 87.5, | |
| "eval_loss": 0.21311460435390472, | |
| "eval_runtime": 43.2323, | |
| "eval_samples_per_second": 27.387, | |
| "eval_steps_per_second": 1.712, | |
| "eval_wer": 0.19309632689187195, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 88.75, | |
| "learning_rate": 9.05e-06, | |
| "loss": 0.5008, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "learning_rate": 8.05e-06, | |
| "loss": 0.4834, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 91.25, | |
| "learning_rate": 7.049999999999999e-06, | |
| "loss": 0.4717, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 92.5, | |
| "learning_rate": 6.049999999999999e-06, | |
| "loss": 0.485, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 93.75, | |
| "learning_rate": 5.05e-06, | |
| "loss": 0.4835, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "learning_rate": 4.049999999999999e-06, | |
| "loss": 0.4787, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 96.25, | |
| "learning_rate": 3.05e-06, | |
| "loss": 0.4747, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 97.5, | |
| "learning_rate": 2.05e-06, | |
| "loss": 0.4804, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 98.75, | |
| "learning_rate": 1.05e-06, | |
| "loss": 0.4731, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 6e-08, | |
| "loss": 0.4797, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_loss": 0.20929841697216034, | |
| "eval_runtime": 42.6795, | |
| "eval_samples_per_second": 27.742, | |
| "eval_steps_per_second": 1.734, | |
| "eval_wer": 0.190736096769435, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 8000, | |
| "total_flos": 2.997146399308124e+19, | |
| "train_loss": 1.1495286922454835, | |
| "train_runtime": 11669.3128, | |
| "train_samples_per_second": 21.921, | |
| "train_steps_per_second": 0.686 | |
| } | |
| ], | |
| "max_steps": 8000, | |
| "num_train_epochs": 100, | |
| "total_flos": 2.997146399308124e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |