| { | |
| "best_metric": 161.45617116397892, | |
| "best_model_checkpoint": "./checkpoint-88", | |
| "epoch": 2.3214285714285716, | |
| "global_step": 112, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0, | |
| "loss": 1.6569, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 2.2058823529411765e-07, | |
| "loss": 1.6564, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.411764705882353e-07, | |
| "loss": 1.6518, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 6.61764705882353e-07, | |
| "loss": 1.6237, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 8.823529411764706e-07, | |
| "loss": 1.7141, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.1029411764705884e-06, | |
| "loss": 1.655, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.323529411764706e-06, | |
| "loss": 1.5904, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.5441176470588234e-06, | |
| "loss": 1.6232, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.7647058823529412e-06, | |
| "loss": 1.6403, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1.985294117647059e-06, | |
| "loss": 1.6065, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 2.2058823529411767e-06, | |
| "loss": 1.5947, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 1.629066824913025, | |
| "eval_runtime": 125.637, | |
| "eval_samples_per_second": 2.627, | |
| "eval_steps_per_second": 0.088, | |
| "eval_wer": 227.94188088775348, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 2.4264705882352943e-06, | |
| "loss": 1.52, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2.647058823529412e-06, | |
| "loss": 1.4456, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2.867647058823529e-06, | |
| "loss": 1.58, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 3.088235294117647e-06, | |
| "loss": 1.5958, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 3.308823529411765e-06, | |
| "loss": 1.4695, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 3.5294117647058825e-06, | |
| "loss": 1.4718, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 3.75e-06, | |
| "loss": 1.4406, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.970588235294118e-06, | |
| "loss": 1.4339, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.191176470588236e-06, | |
| "loss": 1.313, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.411764705882353e-06, | |
| "loss": 1.2825, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.632352941176471e-06, | |
| "loss": 1.3194, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 1.460469126701355, | |
| "eval_runtime": 118.4642, | |
| "eval_samples_per_second": 2.786, | |
| "eval_steps_per_second": 0.093, | |
| "eval_wer": 235.94124221619032, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.852941176470589e-06, | |
| "loss": 1.2301, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 5.073529411764706e-06, | |
| "loss": 1.3231, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 5.294117647058824e-06, | |
| "loss": 1.2689, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 5.5147058823529415e-06, | |
| "loss": 1.2768, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 5.735294117647058e-06, | |
| "loss": 1.2927, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 5.955882352941176e-06, | |
| "loss": 1.3419, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 6.176470588235294e-06, | |
| "loss": 1.141, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 6.397058823529412e-06, | |
| "loss": 1.2907, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 6.61764705882353e-06, | |
| "loss": 1.2248, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 6.838235294117647e-06, | |
| "loss": 1.149, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.058823529411765e-06, | |
| "loss": 1.2636, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.3149287700653076, | |
| "eval_runtime": 115.745, | |
| "eval_samples_per_second": 2.851, | |
| "eval_steps_per_second": 0.095, | |
| "eval_wer": 177.19942519559316, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.2794117647058826e-06, | |
| "loss": 1.2351, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 7.5e-06, | |
| "loss": 1.2055, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 7.403846153846155e-06, | |
| "loss": 1.1864, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 7.307692307692308e-06, | |
| "loss": 1.148, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 7.211538461538462e-06, | |
| "loss": 1.1505, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 7.115384615384615e-06, | |
| "loss": 1.0859, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 7.01923076923077e-06, | |
| "loss": 1.119, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 6.923076923076923e-06, | |
| "loss": 1.0238, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 6.826923076923077e-06, | |
| "loss": 1.0468, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 6.730769230769231e-06, | |
| "loss": 1.061, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 6.6346153846153846e-06, | |
| "loss": 1.0238, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_loss": 1.1836069822311401, | |
| "eval_runtime": 109.1507, | |
| "eval_samples_per_second": 3.023, | |
| "eval_steps_per_second": 0.101, | |
| "eval_wer": 199.80839853105542, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 6.538461538461539e-06, | |
| "loss": 1.0016, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 6.442307692307692e-06, | |
| "loss": 0.9635, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 6.3461538461538466e-06, | |
| "loss": 0.9953, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 6.25e-06, | |
| "loss": 1.0061, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 6.153846153846154e-06, | |
| "loss": 0.9933, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 6.057692307692308e-06, | |
| "loss": 0.9479, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 5.9615384615384615e-06, | |
| "loss": 0.9132, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 5.865384615384615e-06, | |
| "loss": 1.0232, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 5.76923076923077e-06, | |
| "loss": 1.0153, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 5.673076923076923e-06, | |
| "loss": 0.9548, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 5.576923076923077e-06, | |
| "loss": 0.9306, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_loss": 1.1363615989685059, | |
| "eval_runtime": 115.8861, | |
| "eval_samples_per_second": 2.848, | |
| "eval_steps_per_second": 0.095, | |
| "eval_wer": 210.21874501037843, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 5.480769230769231e-06, | |
| "loss": 0.9088, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 5.384615384615385e-06, | |
| "loss": 0.8902, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 5.288461538461539e-06, | |
| "loss": 0.861, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 5.192307692307692e-06, | |
| "loss": 0.8047, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 5.096153846153847e-06, | |
| "loss": 0.8459, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 4.9999999999999996e-06, | |
| "loss": 0.8103, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 4.903846153846154e-06, | |
| "loss": 0.8752, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 4.807692307692308e-06, | |
| "loss": 0.8398, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 4.711538461538462e-06, | |
| "loss": 0.8571, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.8998, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 4.519230769230769e-06, | |
| "loss": 0.9233, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_loss": 1.1090655326843262, | |
| "eval_runtime": 104.9772, | |
| "eval_samples_per_second": 3.144, | |
| "eval_steps_per_second": 0.105, | |
| "eval_wer": 175.20357656075362, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 4.423076923076924e-06, | |
| "loss": 0.7878, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 4.3269230769230765e-06, | |
| "loss": 0.8948, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 4.230769230769231e-06, | |
| "loss": 0.8711, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 4.134615384615385e-06, | |
| "loss": 0.8292, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 4.0384615384615385e-06, | |
| "loss": 0.9318, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 3.942307692307692e-06, | |
| "loss": 0.9272, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 3.846153846153846e-06, | |
| "loss": 0.8923, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3.75e-06, | |
| "loss": 0.9013, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 3.653846153846154e-06, | |
| "loss": 0.8659, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 3.5576923076923075e-06, | |
| "loss": 0.8422, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 3.4615384615384617e-06, | |
| "loss": 0.8482, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 1.0900639295578003, | |
| "eval_runtime": 99.705, | |
| "eval_samples_per_second": 3.31, | |
| "eval_steps_per_second": 0.11, | |
| "eval_wer": 161.85534089094685, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 3.3653846153846154e-06, | |
| "loss": 0.8925, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 3.2692307692307696e-06, | |
| "loss": 0.8202, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 3.1730769230769233e-06, | |
| "loss": 0.8384, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 3.076923076923077e-06, | |
| "loss": 0.8469, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 2.9807692307692307e-06, | |
| "loss": 0.8172, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 2.884615384615385e-06, | |
| "loss": 0.8119, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 2.7884615384615386e-06, | |
| "loss": 0.7811, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 2.6923076923076923e-06, | |
| "loss": 0.819, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 2.596153846153846e-06, | |
| "loss": 0.8324, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 2.4999999999999998e-06, | |
| "loss": 0.8336, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 2.403846153846154e-06, | |
| "loss": 0.7929, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 1.0782241821289062, | |
| "eval_runtime": 111.1236, | |
| "eval_samples_per_second": 2.97, | |
| "eval_steps_per_second": 0.099, | |
| "eval_wer": 161.45617116397892, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 2.307692307692308e-06, | |
| "loss": 0.774, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 2.211538461538462e-06, | |
| "loss": 0.8693, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 2.1153846153846155e-06, | |
| "loss": 0.8628, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 2.0192307692307692e-06, | |
| "loss": 0.8199, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 1.923076923076923e-06, | |
| "loss": 0.7995, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 1.826923076923077e-06, | |
| "loss": 0.7824, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 1.7307692307692308e-06, | |
| "loss": 0.7694, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 1.6346153846153848e-06, | |
| "loss": 0.7526, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1.5384615384615385e-06, | |
| "loss": 0.7037, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 1.4423076923076924e-06, | |
| "loss": 0.7366, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 1.3461538461538462e-06, | |
| "loss": 0.7134, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "eval_loss": 1.0737253427505493, | |
| "eval_runtime": 96.9018, | |
| "eval_samples_per_second": 3.406, | |
| "eval_steps_per_second": 0.114, | |
| "eval_wer": 177.1195912501996, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 1.2499999999999999e-06, | |
| "loss": 0.7738, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 1.153846153846154e-06, | |
| "loss": 0.7373, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 1.0576923076923078e-06, | |
| "loss": 0.7623, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 9.615384615384615e-07, | |
| "loss": 0.815, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 8.653846153846154e-07, | |
| "loss": 0.829, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 7.692307692307693e-07, | |
| "loss": 0.7113, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 6.730769230769231e-07, | |
| "loss": 0.8106, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 5.76923076923077e-07, | |
| "loss": 0.7909, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 4.807692307692307e-07, | |
| "loss": 0.7564, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 3.8461538461538463e-07, | |
| "loss": 0.8574, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 2.884615384615385e-07, | |
| "loss": 0.8543, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_loss": 1.0717276334762573, | |
| "eval_runtime": 95.923, | |
| "eval_samples_per_second": 3.44, | |
| "eval_steps_per_second": 0.115, | |
| "eval_wer": 187.5459045186013, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 1.9230769230769231e-07, | |
| "loss": 0.824, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 9.615384615384616e-08, | |
| "loss": 0.8352, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "step": 112, | |
| "total_flos": 1.7415399333888e+17, | |
| "train_loss": 1.0522634956453527, | |
| "train_runtime": 1735.2585, | |
| "train_samples_per_second": 4.131, | |
| "train_steps_per_second": 0.065 | |
| } | |
| ], | |
| "max_steps": 112, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.7415399333888e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |