| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 150.0, | |
| "global_step": 3450, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.2e-06, | |
| "loss": 20.1322, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1.02e-05, | |
| "loss": 20.2807, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.5599999999999996e-05, | |
| "loss": 19.9863, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 2.1599999999999996e-05, | |
| "loss": 16.998, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 2.7599999999999997e-05, | |
| "loss": 11.753, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 3.36e-05, | |
| "loss": 8.5319, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 3.96e-05, | |
| "loss": 7.0837, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 4.56e-05, | |
| "loss": 6.0396, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 5.1599999999999994e-05, | |
| "loss": 5.5626, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 5.76e-05, | |
| "loss": 5.1697, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 6.359999999999999e-05, | |
| "loss": 4.7468, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 6.96e-05, | |
| "loss": 4.6084, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 7.56e-05, | |
| "loss": 4.0859, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 8.16e-05, | |
| "loss": 3.9953, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 8.759999999999999e-05, | |
| "loss": 3.7522, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 9.36e-05, | |
| "loss": 3.5674, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "learning_rate": 9.96e-05, | |
| "loss": 3.4598, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 0.00010559999999999998, | |
| "loss": 3.42, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "learning_rate": 0.00011159999999999999, | |
| "loss": 3.3579, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 0.0001176, | |
| "loss": 3.3237, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.13, | |
| "learning_rate": 0.0001236, | |
| "loss": 3.2857, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "learning_rate": 0.00012959999999999998, | |
| "loss": 3.2553, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0001356, | |
| "loss": 3.2545, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 10.43, | |
| "learning_rate": 0.00014159999999999997, | |
| "loss": 3.2053, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 0.00014759999999999998, | |
| "loss": 3.249, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 11.3, | |
| "learning_rate": 0.0001536, | |
| "loss": 3.2334, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 11.74, | |
| "learning_rate": 0.0001596, | |
| "loss": 3.2278, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 12.17, | |
| "learning_rate": 0.0001656, | |
| "loss": 3.206, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 12.61, | |
| "learning_rate": 0.00017159999999999997, | |
| "loss": 3.2164, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 0.00017759999999999998, | |
| "loss": 3.2048, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 13.48, | |
| "learning_rate": 0.0001836, | |
| "loss": 3.2024, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 13.91, | |
| "learning_rate": 0.00018959999999999997, | |
| "loss": 3.2113, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 14.35, | |
| "learning_rate": 0.00019559999999999998, | |
| "loss": 3.2023, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 14.78, | |
| "learning_rate": 0.0002016, | |
| "loss": 3.2135, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 15.22, | |
| "learning_rate": 0.00020759999999999998, | |
| "loss": 3.2014, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 15.65, | |
| "learning_rate": 0.00021359999999999996, | |
| "loss": 3.1402, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 16.09, | |
| "learning_rate": 0.00021959999999999997, | |
| "loss": 3.093, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 16.52, | |
| "learning_rate": 0.00022559999999999998, | |
| "loss": 2.9955, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 16.96, | |
| "learning_rate": 0.0002316, | |
| "loss": 2.8186, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 17.39, | |
| "learning_rate": 0.0002376, | |
| "loss": 2.5288, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 17.39, | |
| "eval_loss": 2.0544939041137695, | |
| "eval_runtime": 2.3101, | |
| "eval_samples_per_second": 96.532, | |
| "eval_steps_per_second": 3.03, | |
| "eval_wer": 0.6038511871377827, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 17.83, | |
| "learning_rate": 0.00024359999999999999, | |
| "loss": 2.1609, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 18.26, | |
| "learning_rate": 0.00024959999999999994, | |
| "loss": 1.8764, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 18.7, | |
| "learning_rate": 0.0002556, | |
| "loss": 1.6154, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 19.13, | |
| "learning_rate": 0.00026159999999999996, | |
| "loss": 1.4481, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 19.57, | |
| "learning_rate": 0.0002676, | |
| "loss": 1.2572, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0002736, | |
| "loss": 1.2195, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 20.43, | |
| "learning_rate": 0.00027959999999999997, | |
| "loss": 1.0625, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 20.87, | |
| "learning_rate": 0.00028559999999999995, | |
| "loss": 1.0071, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 21.3, | |
| "learning_rate": 0.0002916, | |
| "loss": 0.9312, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "learning_rate": 0.00029759999999999997, | |
| "loss": 0.8633, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 22.17, | |
| "learning_rate": 0.00029052631578947366, | |
| "loss": 0.8454, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 22.61, | |
| "learning_rate": 0.0002747368421052631, | |
| "loss": 0.7471, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 23.04, | |
| "learning_rate": 0.0002589473684210526, | |
| "loss": 0.7783, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 23.48, | |
| "learning_rate": 0.00024315789473684207, | |
| "loss": 0.64, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 23.91, | |
| "learning_rate": 0.00022736842105263157, | |
| "loss": 0.6077, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 24.35, | |
| "learning_rate": 0.00021157894736842102, | |
| "loss": 0.5887, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 24.78, | |
| "learning_rate": 0.0001957894736842105, | |
| "loss": 0.5389, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 25.22, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 0.4779, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 25.65, | |
| "learning_rate": 0.00016421052631578948, | |
| "loss": 0.4971, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 26.09, | |
| "learning_rate": 0.00014842105263157893, | |
| "loss": 0.4876, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 26.52, | |
| "learning_rate": 0.0001326315789473684, | |
| "loss": 0.4645, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 26.96, | |
| "learning_rate": 0.00011684210526315788, | |
| "loss": 0.3915, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 27.39, | |
| "learning_rate": 0.00010105263157894735, | |
| "loss": 0.3831, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 27.83, | |
| "learning_rate": 8.526315789473684e-05, | |
| "loss": 0.3738, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 28.26, | |
| "learning_rate": 6.947368421052631e-05, | |
| "loss": 0.3274, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 28.7, | |
| "learning_rate": 5.3684210526315784e-05, | |
| "loss": 0.3508, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 29.13, | |
| "learning_rate": 3.789473684210526e-05, | |
| "loss": 0.3293, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 29.57, | |
| "learning_rate": 2.2105263157894733e-05, | |
| "loss": 0.3146, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 6.3157894736842095e-06, | |
| "loss": 0.2893, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "step": 690, | |
| "total_flos": 3.7988686760446525e+18, | |
| "train_loss": 3.560509018276049, | |
| "train_runtime": 533.4671, | |
| "train_samples_per_second": 40.49, | |
| "train_steps_per_second": 1.293 | |
| }, | |
| { | |
| "epoch": 30.43, | |
| "learning_rate": 5.999999999999999e-06, | |
| "loss": 0.3166, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 30.87, | |
| "learning_rate": 1.1999999999999999e-05, | |
| "loss": 0.3066, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 31.3, | |
| "learning_rate": 1.7999999999999997e-05, | |
| "loss": 0.2917, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 31.74, | |
| "learning_rate": 2.3999999999999997e-05, | |
| "loss": 0.2866, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 32.17, | |
| "learning_rate": 2.9999999999999997e-05, | |
| "loss": 0.2835, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 32.61, | |
| "learning_rate": 3.5999999999999994e-05, | |
| "loss": 0.2765, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 33.04, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.2951, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 33.48, | |
| "learning_rate": 4.7999999999999994e-05, | |
| "loss": 0.2685, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 33.91, | |
| "learning_rate": 5.399999999999999e-05, | |
| "loss": 0.3011, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 34.35, | |
| "learning_rate": 5.94e-05, | |
| "loss": 0.273, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 34.78, | |
| "learning_rate": 6.479999999999999e-05, | |
| "loss": 0.2616, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 34.78, | |
| "eval_loss": 0.4040215313434601, | |
| "eval_runtime": 2.307, | |
| "eval_samples_per_second": 96.662, | |
| "eval_steps_per_second": 3.034, | |
| "eval_wer": 0.12422882781828379, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 35.22, | |
| "learning_rate": 7.079999999999999e-05, | |
| "loss": 0.2468, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 35.65, | |
| "learning_rate": 7.68e-05, | |
| "loss": 0.2571, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 36.09, | |
| "learning_rate": 8.28e-05, | |
| "loss": 0.2412, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 36.52, | |
| "learning_rate": 8.879999999999999e-05, | |
| "loss": 0.2392, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 36.96, | |
| "learning_rate": 9.479999999999999e-05, | |
| "loss": 0.2594, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 37.39, | |
| "learning_rate": 0.0001008, | |
| "loss": 0.2413, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 37.83, | |
| "learning_rate": 0.00010679999999999998, | |
| "loss": 0.2502, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 38.26, | |
| "learning_rate": 0.00011279999999999999, | |
| "loss": 0.248, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 38.7, | |
| "learning_rate": 0.0001188, | |
| "loss": 0.2272, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 39.13, | |
| "learning_rate": 0.00012479999999999997, | |
| "loss": 0.2359, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 39.57, | |
| "learning_rate": 0.00013079999999999998, | |
| "loss": 0.2299, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 0.0001368, | |
| "loss": 0.2321, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 40.43, | |
| "learning_rate": 0.00014279999999999997, | |
| "loss": 0.2385, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 40.87, | |
| "learning_rate": 0.00014879999999999998, | |
| "loss": 0.2183, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 41.3, | |
| "learning_rate": 0.0001548, | |
| "loss": 0.2318, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 41.74, | |
| "learning_rate": 0.0001608, | |
| "loss": 0.241, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 42.17, | |
| "learning_rate": 0.0001668, | |
| "loss": 0.2314, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 42.61, | |
| "learning_rate": 0.00017279999999999997, | |
| "loss": 0.2195, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 43.04, | |
| "learning_rate": 0.00017879999999999998, | |
| "loss": 0.232, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 43.48, | |
| "learning_rate": 0.0001848, | |
| "loss": 0.196, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 43.91, | |
| "learning_rate": 0.00019079999999999998, | |
| "loss": 0.2384, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 44.35, | |
| "learning_rate": 0.00019679999999999999, | |
| "loss": 0.2293, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 44.78, | |
| "learning_rate": 0.0002028, | |
| "loss": 0.2241, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 45.22, | |
| "learning_rate": 0.00020879999999999998, | |
| "loss": 0.1895, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 45.65, | |
| "learning_rate": 0.00021479999999999996, | |
| "loss": 0.2248, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 46.09, | |
| "learning_rate": 0.00022079999999999997, | |
| "loss": 0.2288, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 46.52, | |
| "learning_rate": 0.00022679999999999998, | |
| "loss": 0.2115, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 46.96, | |
| "learning_rate": 0.0002328, | |
| "loss": 0.2041, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 47.39, | |
| "learning_rate": 0.0002388, | |
| "loss": 0.2281, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 47.83, | |
| "learning_rate": 0.0002448, | |
| "loss": 0.2306, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 48.26, | |
| "learning_rate": 0.00025079999999999997, | |
| "loss": 0.2039, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 48.7, | |
| "learning_rate": 0.00025679999999999995, | |
| "loss": 0.2167, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 49.13, | |
| "learning_rate": 0.0002628, | |
| "loss": 0.2339, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 49.57, | |
| "learning_rate": 0.0002688, | |
| "loss": 0.2366, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 0.0002748, | |
| "loss": 0.1972, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 1150, | |
| "total_flos": 6.332515423493292e+18, | |
| "train_loss": 0.09717807769775391, | |
| "train_runtime": 228.6677, | |
| "train_samples_per_second": 157.434, | |
| "train_steps_per_second": 5.029 | |
| }, | |
| { | |
| "epoch": 52.17, | |
| "learning_rate": 3.675e-06, | |
| "loss": 1.0586, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 56.52, | |
| "learning_rate": 1.1099999999999999e-05, | |
| "loss": 0.9511, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 60.87, | |
| "learning_rate": 1.8599999999999998e-05, | |
| "loss": 0.8827, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 65.22, | |
| "learning_rate": 2.6024999999999996e-05, | |
| "loss": 0.8123, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 65.22, | |
| "eval_loss": 0.38979411125183105, | |
| "eval_runtime": 2.5496, | |
| "eval_samples_per_second": 87.465, | |
| "eval_steps_per_second": 2.746, | |
| "eval_wer": 0.11226397457468686, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 69.57, | |
| "learning_rate": 3.3524999999999995e-05, | |
| "loss": 0.7657, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 73.91, | |
| "learning_rate": 4.1025e-05, | |
| "loss": 0.7197, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 78.26, | |
| "learning_rate": 4.845e-05, | |
| "loss": 0.6831, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 82.61, | |
| "learning_rate": 5.595e-05, | |
| "loss": 0.6424, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 86.96, | |
| "learning_rate": 6.345e-05, | |
| "loss": 0.6314, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 86.96, | |
| "eval_loss": 0.3785865306854248, | |
| "eval_runtime": 2.183, | |
| "eval_samples_per_second": 102.155, | |
| "eval_steps_per_second": 3.207, | |
| "eval_wer": 0.10431856421761077, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 91.3, | |
| "learning_rate": 7.094999999999999e-05, | |
| "loss": 0.5952, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 95.65, | |
| "learning_rate": 7.359183673469387e-05, | |
| "loss": 0.5769, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 7.053061224489794e-05, | |
| "loss": 0.5631, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 104.35, | |
| "learning_rate": 6.746938775510203e-05, | |
| "loss": 0.5318, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 108.7, | |
| "learning_rate": 6.440816326530611e-05, | |
| "loss": 0.5046, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 108.7, | |
| "eval_loss": 0.38604071736335754, | |
| "eval_runtime": 2.1232, | |
| "eval_samples_per_second": 105.03, | |
| "eval_steps_per_second": 3.297, | |
| "eval_wer": 0.10515984296130118, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 113.04, | |
| "learning_rate": 6.13469387755102e-05, | |
| "loss": 0.5035, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 117.39, | |
| "learning_rate": 5.828571428571428e-05, | |
| "loss": 0.4646, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 121.74, | |
| "learning_rate": 5.525510204081632e-05, | |
| "loss": 0.4552, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 126.09, | |
| "learning_rate": 5.21938775510204e-05, | |
| "loss": 0.4378, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 130.43, | |
| "learning_rate": 4.9132653061224486e-05, | |
| "loss": 0.4264, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 130.43, | |
| "eval_loss": 0.37559226155281067, | |
| "eval_runtime": 2.1713, | |
| "eval_samples_per_second": 102.705, | |
| "eval_steps_per_second": 3.224, | |
| "eval_wer": 0.10001869508319312, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 134.78, | |
| "learning_rate": 4.607142857142857e-05, | |
| "loss": 0.4114, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 139.13, | |
| "learning_rate": 4.3010204081632646e-05, | |
| "loss": 0.3945, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 143.48, | |
| "learning_rate": 3.994897959183673e-05, | |
| "loss": 0.3842, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 147.83, | |
| "learning_rate": 3.688775510204081e-05, | |
| "loss": 0.3757, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 150.0, | |
| "step": 3450, | |
| "total_flos": 1.8994436456459534e+19, | |
| "train_loss": 0.38979322350543477, | |
| "train_runtime": 1100.4507, | |
| "train_samples_per_second": 98.142, | |
| "train_steps_per_second": 3.135 | |
| } | |
| ], | |
| "max_steps": 3450, | |
| "num_train_epochs": 150, | |
| "total_flos": 1.8994436456459534e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |