| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 256.4102564102564, | |
| "global_step": 20000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 0.000499, | |
| "loss": 1.4757, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "eval_loss": 0.061434268951416016, | |
| "eval_runtime": 68.9385, | |
| "eval_samples_per_second": 26.937, | |
| "eval_steps_per_second": 1.131, | |
| "eval_wer": 0.03469206911829863, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 12.82, | |
| "learning_rate": 0.0004872051282051282, | |
| "loss": 0.0624, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 12.82, | |
| "eval_loss": 0.05250529572367668, | |
| "eval_runtime": 68.4693, | |
| "eval_samples_per_second": 27.122, | |
| "eval_steps_per_second": 1.139, | |
| "eval_wer": 0.027735932653965442, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "learning_rate": 0.0004743846153846154, | |
| "loss": 0.0388, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "eval_loss": 0.06932416558265686, | |
| "eval_runtime": 67.8311, | |
| "eval_samples_per_second": 27.377, | |
| "eval_steps_per_second": 1.15, | |
| "eval_wer": 0.024147097917589722, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 25.64, | |
| "learning_rate": 0.0004615641025641026, | |
| "loss": 0.03, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 25.64, | |
| "eval_loss": 0.0665554478764534, | |
| "eval_runtime": 69.0525, | |
| "eval_samples_per_second": 26.893, | |
| "eval_steps_per_second": 1.13, | |
| "eval_wer": 0.024412937527691626, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 32.05, | |
| "learning_rate": 0.0004487435897435898, | |
| "loss": 0.0235, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 32.05, | |
| "eval_loss": 0.06039978191256523, | |
| "eval_runtime": 67.8702, | |
| "eval_samples_per_second": 27.361, | |
| "eval_steps_per_second": 1.149, | |
| "eval_wer": 0.026007975188303056, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "learning_rate": 0.0004359230769230769, | |
| "loss": 0.0226, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "eval_loss": 0.06249115616083145, | |
| "eval_runtime": 67.3401, | |
| "eval_samples_per_second": 27.576, | |
| "eval_steps_per_second": 1.158, | |
| "eval_wer": 0.022950819672131147, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 44.87, | |
| "learning_rate": 0.0004231025641025641, | |
| "loss": 0.0163, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 44.87, | |
| "eval_loss": 0.06026700139045715, | |
| "eval_runtime": 67.7207, | |
| "eval_samples_per_second": 27.421, | |
| "eval_steps_per_second": 1.152, | |
| "eval_wer": 0.01953921134249003, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 51.28, | |
| "learning_rate": 0.0004102820512820513, | |
| "loss": 0.0157, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 51.28, | |
| "eval_loss": 0.0627710148692131, | |
| "eval_runtime": 67.8528, | |
| "eval_samples_per_second": 27.368, | |
| "eval_steps_per_second": 1.15, | |
| "eval_wer": 0.02091271599468321, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 57.69, | |
| "learning_rate": 0.0003974615384615385, | |
| "loss": 0.0152, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 57.69, | |
| "eval_loss": 0.06921559572219849, | |
| "eval_runtime": 67.9506, | |
| "eval_samples_per_second": 27.329, | |
| "eval_steps_per_second": 1.148, | |
| "eval_wer": 0.023836951705804164, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 64.1, | |
| "learning_rate": 0.00038464102564102567, | |
| "loss": 0.0122, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 64.1, | |
| "eval_loss": 0.06069288030266762, | |
| "eval_runtime": 67.6564, | |
| "eval_samples_per_second": 27.448, | |
| "eval_steps_per_second": 1.153, | |
| "eval_wer": 0.02100132919805051, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 70.51, | |
| "learning_rate": 0.0003718205128205128, | |
| "loss": 0.011, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 70.51, | |
| "eval_loss": 0.06082445755600929, | |
| "eval_runtime": 68.2551, | |
| "eval_samples_per_second": 27.207, | |
| "eval_steps_per_second": 1.143, | |
| "eval_wer": 0.021311475409836064, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 76.92, | |
| "learning_rate": 0.000359, | |
| "loss": 0.0114, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 76.92, | |
| "eval_loss": 0.06809388101100922, | |
| "eval_runtime": 68.8171, | |
| "eval_samples_per_second": 26.985, | |
| "eval_steps_per_second": 1.133, | |
| "eval_wer": 0.021089942401417813, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 83.33, | |
| "learning_rate": 0.00034620512820512825, | |
| "loss": 0.0106, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 83.33, | |
| "eval_loss": 0.061287231743335724, | |
| "eval_runtime": 67.6622, | |
| "eval_samples_per_second": 27.445, | |
| "eval_steps_per_second": 1.153, | |
| "eval_wer": 0.02100132919805051, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 89.74, | |
| "learning_rate": 0.0003333846153846154, | |
| "loss": 0.0081, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 89.74, | |
| "eval_loss": 0.06544984877109528, | |
| "eval_runtime": 68.1647, | |
| "eval_samples_per_second": 27.243, | |
| "eval_steps_per_second": 1.144, | |
| "eval_wer": 0.019627824545857334, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 96.15, | |
| "learning_rate": 0.00032056410256410257, | |
| "loss": 0.0078, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 96.15, | |
| "eval_loss": 0.06121820956468582, | |
| "eval_runtime": 68.399, | |
| "eval_samples_per_second": 27.15, | |
| "eval_steps_per_second": 1.14, | |
| "eval_wer": 0.01905183872396987, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 102.56, | |
| "learning_rate": 0.00030774358974358976, | |
| "loss": 0.0082, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 102.56, | |
| "eval_loss": 0.07580074667930603, | |
| "eval_runtime": 67.4807, | |
| "eval_samples_per_second": 27.519, | |
| "eval_steps_per_second": 1.156, | |
| "eval_wer": 0.02365972529906956, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 108.97, | |
| "learning_rate": 0.0002949230769230769, | |
| "loss": 0.0078, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 108.97, | |
| "eval_loss": 0.06641749292612076, | |
| "eval_runtime": 67.4225, | |
| "eval_samples_per_second": 27.543, | |
| "eval_steps_per_second": 1.157, | |
| "eval_wer": 0.02060256978289765, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 115.38, | |
| "learning_rate": 0.00028210256410256414, | |
| "loss": 0.0075, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 115.38, | |
| "eval_loss": 0.06580852717161179, | |
| "eval_runtime": 67.9347, | |
| "eval_samples_per_second": 27.335, | |
| "eval_steps_per_second": 1.148, | |
| "eval_wer": 0.019716437749224634, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 121.79, | |
| "learning_rate": 0.0002692820512820513, | |
| "loss": 0.0052, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 121.79, | |
| "eval_loss": 0.06690431386232376, | |
| "eval_runtime": 66.8582, | |
| "eval_samples_per_second": 27.775, | |
| "eval_steps_per_second": 1.167, | |
| "eval_wer": 0.021843154630039874, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 128.21, | |
| "learning_rate": 0.00025646153846153847, | |
| "loss": 0.0054, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 128.21, | |
| "eval_loss": 0.06948971748352051, | |
| "eval_runtime": 67.1148, | |
| "eval_samples_per_second": 27.669, | |
| "eval_steps_per_second": 1.162, | |
| "eval_wer": 0.021089942401417813, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 134.62, | |
| "learning_rate": 0.00024364102564102563, | |
| "loss": 0.0053, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 134.62, | |
| "eval_loss": 0.07259159535169601, | |
| "eval_runtime": 66.7066, | |
| "eval_samples_per_second": 27.838, | |
| "eval_steps_per_second": 1.169, | |
| "eval_wer": 0.022684980062029243, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 141.03, | |
| "learning_rate": 0.00023082051282051282, | |
| "loss": 0.0046, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 141.03, | |
| "eval_loss": 0.07016939669847488, | |
| "eval_runtime": 68.2063, | |
| "eval_samples_per_second": 27.226, | |
| "eval_steps_per_second": 1.144, | |
| "eval_wer": 0.021178555604785113, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 147.44, | |
| "learning_rate": 0.000218, | |
| "loss": 0.0043, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 147.44, | |
| "eval_loss": 0.08461024612188339, | |
| "eval_runtime": 66.3999, | |
| "eval_samples_per_second": 27.967, | |
| "eval_steps_per_second": 1.175, | |
| "eval_wer": 0.020026583961010192, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 153.85, | |
| "learning_rate": 0.00020517948717948718, | |
| "loss": 0.0041, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 153.85, | |
| "eval_loss": 0.07643292099237442, | |
| "eval_runtime": 66.6168, | |
| "eval_samples_per_second": 27.876, | |
| "eval_steps_per_second": 1.171, | |
| "eval_wer": 0.020026583961010192, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 160.26, | |
| "learning_rate": 0.00019235897435897437, | |
| "loss": 0.0032, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 160.26, | |
| "eval_loss": 0.0785411074757576, | |
| "eval_runtime": 67.7953, | |
| "eval_samples_per_second": 27.391, | |
| "eval_steps_per_second": 1.151, | |
| "eval_wer": 0.02007089056269384, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 166.67, | |
| "learning_rate": 0.0001795641025641026, | |
| "loss": 0.0028, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 166.67, | |
| "eval_loss": 0.08392436057329178, | |
| "eval_runtime": 67.1052, | |
| "eval_samples_per_second": 27.673, | |
| "eval_steps_per_second": 1.162, | |
| "eval_wer": 0.019672131147540985, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 173.08, | |
| "learning_rate": 0.00016674358974358975, | |
| "loss": 0.0035, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 173.08, | |
| "eval_loss": 0.07846853882074356, | |
| "eval_runtime": 66.3462, | |
| "eval_samples_per_second": 27.99, | |
| "eval_steps_per_second": 1.176, | |
| "eval_wer": 0.02100132919805051, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 179.49, | |
| "learning_rate": 0.00015394871794871794, | |
| "loss": 0.0027, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 179.49, | |
| "eval_loss": 0.07303400337696075, | |
| "eval_runtime": 66.6716, | |
| "eval_samples_per_second": 27.853, | |
| "eval_steps_per_second": 1.17, | |
| "eval_wer": 0.018785999113867965, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 185.9, | |
| "learning_rate": 0.00014112820512820513, | |
| "loss": 0.002, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 185.9, | |
| "eval_loss": 0.07940459251403809, | |
| "eval_runtime": 66.4899, | |
| "eval_samples_per_second": 27.929, | |
| "eval_steps_per_second": 1.173, | |
| "eval_wer": 0.019317678334071775, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 192.31, | |
| "learning_rate": 0.00012830769230769232, | |
| "loss": 0.002, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 192.31, | |
| "eval_loss": 0.08587377518415451, | |
| "eval_runtime": 67.8686, | |
| "eval_samples_per_second": 27.362, | |
| "eval_steps_per_second": 1.149, | |
| "eval_wer": 0.02113424900310146, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 198.72, | |
| "learning_rate": 0.00011548717948717949, | |
| "loss": 0.0019, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 198.72, | |
| "eval_loss": 0.07269652187824249, | |
| "eval_runtime": 66.1737, | |
| "eval_samples_per_second": 28.063, | |
| "eval_steps_per_second": 1.179, | |
| "eval_wer": 0.018342933097031458, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 205.13, | |
| "learning_rate": 0.00010266666666666668, | |
| "loss": 0.0017, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 205.13, | |
| "eval_loss": 0.07843895256519318, | |
| "eval_runtime": 67.2939, | |
| "eval_samples_per_second": 27.595, | |
| "eval_steps_per_second": 1.159, | |
| "eval_wer": 0.018653079308817013, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 211.54, | |
| "learning_rate": 8.984615384615384e-05, | |
| "loss": 0.0016, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 211.54, | |
| "eval_loss": 0.08008446544408798, | |
| "eval_runtime": 67.2442, | |
| "eval_samples_per_second": 27.616, | |
| "eval_steps_per_second": 1.16, | |
| "eval_wer": 0.019627824545857334, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 217.95, | |
| "learning_rate": 7.702564102564103e-05, | |
| "loss": 0.0014, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 217.95, | |
| "eval_loss": 0.0820729061961174, | |
| "eval_runtime": 67.2612, | |
| "eval_samples_per_second": 27.609, | |
| "eval_steps_per_second": 1.16, | |
| "eval_wer": 0.01847585290208241, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 224.36, | |
| "learning_rate": 6.420512820512821e-05, | |
| "loss": 0.0011, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 224.36, | |
| "eval_loss": 0.08215450495481491, | |
| "eval_runtime": 67.1514, | |
| "eval_samples_per_second": 27.654, | |
| "eval_steps_per_second": 1.162, | |
| "eval_wer": 0.017634027470093044, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 230.77, | |
| "learning_rate": 5.1384615384615385e-05, | |
| "loss": 0.001, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 230.77, | |
| "eval_loss": 0.0855555310845375, | |
| "eval_runtime": 67.2619, | |
| "eval_samples_per_second": 27.609, | |
| "eval_steps_per_second": 1.16, | |
| "eval_wer": 0.017058041648205582, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 237.18, | |
| "learning_rate": 3.856410256410257e-05, | |
| "loss": 0.001, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 237.18, | |
| "eval_loss": 0.07919255644083023, | |
| "eval_runtime": 66.733, | |
| "eval_samples_per_second": 27.827, | |
| "eval_steps_per_second": 1.169, | |
| "eval_wer": 0.017589720868409393, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 243.59, | |
| "learning_rate": 2.574358974358974e-05, | |
| "loss": 0.001, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 243.59, | |
| "eval_loss": 0.08259344846010208, | |
| "eval_runtime": 67.3537, | |
| "eval_samples_per_second": 27.571, | |
| "eval_steps_per_second": 1.158, | |
| "eval_wer": 0.017279574656623838, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 250.0, | |
| "learning_rate": 1.2923076923076924e-05, | |
| "loss": 0.0006, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 250.0, | |
| "eval_loss": 0.08542538434267044, | |
| "eval_runtime": 66.6818, | |
| "eval_samples_per_second": 27.849, | |
| "eval_steps_per_second": 1.17, | |
| "eval_wer": 0.016969428444838282, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 256.41, | |
| "learning_rate": 1.282051282051282e-07, | |
| "loss": 0.0007, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 256.41, | |
| "eval_loss": 0.08499366790056229, | |
| "eval_runtime": 66.4242, | |
| "eval_samples_per_second": 27.957, | |
| "eval_steps_per_second": 1.174, | |
| "eval_wer": 0.016747895436420027, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 256.41, | |
| "step": 20000, | |
| "total_flos": 2.406420737737408e+20, | |
| "train_loss": 0.04605116495639086, | |
| "train_runtime": 87629.7749, | |
| "train_samples_per_second": 21.91, | |
| "train_steps_per_second": 0.228 | |
| } | |
| ], | |
| "max_steps": 20000, | |
| "num_train_epochs": 257, | |
| "total_flos": 2.406420737737408e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |