| { | |
| "best_metric": 158.1031454574485, | |
| "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": 4.347826086956522e-07, | |
| "loss": 1.6564, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 8.695652173913044e-07, | |
| "loss": 1.6518, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1.3043478260869566e-06, | |
| "loss": 1.6233, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1.7391304347826088e-06, | |
| "loss": 1.712, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 2.173913043478261e-06, | |
| "loss": 1.6476, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.6086956521739132e-06, | |
| "loss": 1.5716, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 3.043478260869566e-06, | |
| "loss": 1.5909, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 3.4782608695652175e-06, | |
| "loss": 1.5905, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 3.91304347826087e-06, | |
| "loss": 1.5486, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.347826086956522e-06, | |
| "loss": 1.5299, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 1.5621598958969116, | |
| "eval_runtime": 105.0814, | |
| "eval_samples_per_second": 3.14, | |
| "eval_steps_per_second": 0.105, | |
| "eval_wer": 219.67108414497844, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.782608695652174e-06, | |
| "loss": 1.4537, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 5.2173913043478265e-06, | |
| "loss": 1.3729, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 5.652173913043479e-06, | |
| "loss": 1.4921, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 6.086956521739132e-06, | |
| "loss": 1.5069, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 6.521739130434783e-06, | |
| "loss": 1.388, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 6.956521739130435e-06, | |
| "loss": 1.3857, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 7.391304347826087e-06, | |
| "loss": 1.3389, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 7.82608695652174e-06, | |
| "loss": 1.3089, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 8.260869565217392e-06, | |
| "loss": 1.2118, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 8.695652173913044e-06, | |
| "loss": 1.1634, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.130434782608697e-06, | |
| "loss": 1.1908, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 1.3651723861694336, | |
| "eval_runtime": 110.8115, | |
| "eval_samples_per_second": 2.978, | |
| "eval_steps_per_second": 0.099, | |
| "eval_wer": 192.2401405077439, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 9.565217391304349e-06, | |
| "loss": 1.1192, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2042, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 9.887640449438202e-06, | |
| "loss": 1.1448, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 9.775280898876405e-06, | |
| "loss": 1.1393, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 9.662921348314608e-06, | |
| "loss": 1.1482, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 9.55056179775281e-06, | |
| "loss": 1.179, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 9.438202247191012e-06, | |
| "loss": 0.9847, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 9.325842696629213e-06, | |
| "loss": 1.1149, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 9.213483146067417e-06, | |
| "loss": 1.0657, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 9.101123595505619e-06, | |
| "loss": 1.0028, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 8.988764044943822e-06, | |
| "loss": 1.1161, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.1921106576919556, | |
| "eval_runtime": 103.0628, | |
| "eval_samples_per_second": 3.202, | |
| "eval_steps_per_second": 0.107, | |
| "eval_wer": 200.23950183618075, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 8.876404494382023e-06, | |
| "loss": 1.0931, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 8.764044943820226e-06, | |
| "loss": 1.0635, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 8.651685393258428e-06, | |
| "loss": 1.0671, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 8.53932584269663e-06, | |
| "loss": 1.0251, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 8.426966292134832e-06, | |
| "loss": 1.0486, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 8.314606741573035e-06, | |
| "loss": 0.9926, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 8.202247191011237e-06, | |
| "loss": 1.0232, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 8.08988764044944e-06, | |
| "loss": 0.9421, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 7.97752808988764e-06, | |
| "loss": 0.9541, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 7.865168539325843e-06, | |
| "loss": 0.9639, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 7.752808988764046e-06, | |
| "loss": 0.9216, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_loss": 1.1263455152511597, | |
| "eval_runtime": 93.6663, | |
| "eval_samples_per_second": 3.523, | |
| "eval_steps_per_second": 0.117, | |
| "eval_wer": 186.52403001756346, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 7.640449438202247e-06, | |
| "loss": 0.9097, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 7.5280898876404495e-06, | |
| "loss": 0.8688, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 7.415730337078652e-06, | |
| "loss": 0.9019, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 7.303370786516854e-06, | |
| "loss": 0.9135, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 7.191011235955056e-06, | |
| "loss": 0.9033, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 7.078651685393258e-06, | |
| "loss": 0.8575, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 6.966292134831461e-06, | |
| "loss": 0.8276, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 6.853932584269663e-06, | |
| "loss": 0.9276, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 6.741573033707865e-06, | |
| "loss": 0.9186, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 6.629213483146067e-06, | |
| "loss": 0.8693, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 6.51685393258427e-06, | |
| "loss": 0.8441, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_loss": 1.0945535898208618, | |
| "eval_runtime": 102.0356, | |
| "eval_samples_per_second": 3.234, | |
| "eval_steps_per_second": 0.108, | |
| "eval_wer": 179.32300814306242, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 6.404494382022472e-06, | |
| "loss": 0.8189, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 6.292134831460674e-06, | |
| "loss": 0.8047, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 6.179775280898876e-06, | |
| "loss": 0.7858, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 6.06741573033708e-06, | |
| "loss": 0.7312, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 5.955056179775281e-06, | |
| "loss": 0.7638, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 5.842696629213483e-06, | |
| "loss": 0.7374, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 5.730337078651685e-06, | |
| "loss": 0.7986, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 5.617977528089889e-06, | |
| "loss": 0.762, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 5.50561797752809e-06, | |
| "loss": 0.785, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 5.393258426966292e-06, | |
| "loss": 0.8349, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 5.280898876404494e-06, | |
| "loss": 0.8505, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_loss": 1.0748353004455566, | |
| "eval_runtime": 93.3893, | |
| "eval_samples_per_second": 3.534, | |
| "eval_steps_per_second": 0.118, | |
| "eval_wer": 159.68385757624142, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 5.168539325842698e-06, | |
| "loss": 0.7309, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 5.0561797752809e-06, | |
| "loss": 0.8261, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 4.943820224719101e-06, | |
| "loss": 0.8053, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 4.831460674157304e-06, | |
| "loss": 0.7672, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 4.719101123595506e-06, | |
| "loss": 0.8692, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 4.606741573033709e-06, | |
| "loss": 0.8588, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 4.494382022471911e-06, | |
| "loss": 0.8277, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 4.382022471910113e-06, | |
| "loss": 0.8387, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 4.269662921348315e-06, | |
| "loss": 0.801, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 4.157303370786518e-06, | |
| "loss": 0.7801, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 4.04494382022472e-06, | |
| "loss": 0.7844, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 1.0585265159606934, | |
| "eval_runtime": 87.6428, | |
| "eval_samples_per_second": 3.765, | |
| "eval_steps_per_second": 0.126, | |
| "eval_wer": 163.2923519080313, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 3.932584269662922e-06, | |
| "loss": 0.8227, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 3.820224719101124e-06, | |
| "loss": 0.757, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 3.707865168539326e-06, | |
| "loss": 0.7713, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 3.595505617977528e-06, | |
| "loss": 0.7782, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 3.4831460674157306e-06, | |
| "loss": 0.7468, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 3.3707865168539327e-06, | |
| "loss": 0.7457, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 3.258426966292135e-06, | |
| "loss": 0.7126, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 3.146067415730337e-06, | |
| "loss": 0.7476, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 3.03370786516854e-06, | |
| "loss": 0.76, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 2.9213483146067416e-06, | |
| "loss": 0.7673, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 2.8089887640449444e-06, | |
| "loss": 0.7208, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 1.0490810871124268, | |
| "eval_runtime": 106.8845, | |
| "eval_samples_per_second": 3.087, | |
| "eval_steps_per_second": 0.103, | |
| "eval_wer": 158.1031454574485, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 2.696629213483146e-06, | |
| "loss": 0.7045, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 2.584269662921349e-06, | |
| "loss": 0.7887, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 2.4719101123595505e-06, | |
| "loss": 0.7821, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 2.359550561797753e-06, | |
| "loss": 0.7487, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 2.2471910112359554e-06, | |
| "loss": 0.7281, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 2.1348314606741574e-06, | |
| "loss": 0.7084, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 2.02247191011236e-06, | |
| "loss": 0.6971, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 1.910112359550562e-06, | |
| "loss": 0.6869, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1.797752808988764e-06, | |
| "loss": 0.6411, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 1.6853932584269663e-06, | |
| "loss": 0.6658, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 1.5730337078651686e-06, | |
| "loss": 0.6481, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "eval_loss": 1.046801209449768, | |
| "eval_runtime": 89.0792, | |
| "eval_samples_per_second": 3.705, | |
| "eval_steps_per_second": 0.123, | |
| "eval_wer": 158.51828197349514, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 1.4606741573033708e-06, | |
| "loss": 0.7064, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 1.348314606741573e-06, | |
| "loss": 0.6699, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 1.2359550561797752e-06, | |
| "loss": 0.6974, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 1.1235955056179777e-06, | |
| "loss": 0.7523, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 1.01123595505618e-06, | |
| "loss": 0.7621, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 8.98876404494382e-07, | |
| "loss": 0.657, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 7.865168539325843e-07, | |
| "loss": 0.7443, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 6.741573033707865e-07, | |
| "loss": 0.729, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 5.617977528089888e-07, | |
| "loss": 0.6983, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 4.49438202247191e-07, | |
| "loss": 0.7963, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 3.3707865168539325e-07, | |
| "loss": 0.7912, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_loss": 1.045613408088684, | |
| "eval_runtime": 91.8976, | |
| "eval_samples_per_second": 3.591, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 168.6092926712438, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 2.247191011235955e-07, | |
| "loss": 0.7626, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 1.1235955056179776e-07, | |
| "loss": 0.7754, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "step": 112, | |
| "total_flos": 1.7415399333888e+17, | |
| "train_loss": 0.9721650715385165, | |
| "train_runtime": 1491.7987, | |
| "train_samples_per_second": 4.805, | |
| "train_steps_per_second": 0.075 | |
| } | |
| ], | |
| "max_steps": 112, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.7415399333888e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |