| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 99.99749373433583, | |
| "global_step": 19900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 3.6375e-06, | |
| "loss": 136.5086, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 7.3875e-06, | |
| "loss": 98.7644, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 1.1137499999999998e-05, | |
| "loss": 81.5207, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 1.48875e-05, | |
| "loss": 72.7656, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 1.86375e-05, | |
| "loss": 64.6189, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "eval_cer": 1.0, | |
| "eval_loss": 63.80772399902344, | |
| "eval_runtime": 134.6445, | |
| "eval_samples_per_second": 21.969, | |
| "eval_steps_per_second": 2.748, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 2.23875e-05, | |
| "loss": 53.9323, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 2.6137499999999995e-05, | |
| "loss": 41.1416, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 2.9887499999999998e-05, | |
| "loss": 27.1675, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 3.36375e-05, | |
| "loss": 14.6933, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 3.7387499999999994e-05, | |
| "loss": 8.0561, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "eval_cer": 1.0, | |
| "eval_loss": 6.8014140129089355, | |
| "eval_runtime": 123.7155, | |
| "eval_samples_per_second": 23.91, | |
| "eval_steps_per_second": 2.991, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 4.11375e-05, | |
| "loss": 6.5122, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 4.48875e-05, | |
| "loss": 6.1973, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 4.8637499999999996e-05, | |
| "loss": 6.0687, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 5.23875e-05, | |
| "loss": 6.0963, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "learning_rate": 5.61375e-05, | |
| "loss": 6.0427, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "eval_cer": 1.0, | |
| "eval_loss": 6.074538230895996, | |
| "eval_runtime": 122.7663, | |
| "eval_samples_per_second": 24.095, | |
| "eval_steps_per_second": 3.014, | |
| "eval_wer": 1.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 5.988749999999999e-05, | |
| "loss": 6.0461, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 8.54, | |
| "learning_rate": 6.36375e-05, | |
| "loss": 5.9929, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 6.738749999999999e-05, | |
| "loss": 6.0191, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 9.55, | |
| "learning_rate": 7.11375e-05, | |
| "loss": 5.938, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 7.48875e-05, | |
| "loss": 5.9357, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "eval_cer": 1.0, | |
| "eval_loss": 5.868249416351318, | |
| "eval_runtime": 121.8577, | |
| "eval_samples_per_second": 24.274, | |
| "eval_steps_per_second": 3.036, | |
| "eval_wer": 1.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 7.45935754189944e-05, | |
| "loss": 5.8143, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 7.417458100558659e-05, | |
| "loss": 5.7549, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "learning_rate": 7.375558659217876e-05, | |
| "loss": 5.5454, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 7.333659217877095e-05, | |
| "loss": 5.3124, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 12.56, | |
| "learning_rate": 7.291759776536312e-05, | |
| "loss": 5.0489, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 12.56, | |
| "eval_cer": 0.7749867254829882, | |
| "eval_loss": 4.4031829833984375, | |
| "eval_runtime": 123.6444, | |
| "eval_samples_per_second": 23.923, | |
| "eval_steps_per_second": 2.992, | |
| "eval_wer": 0.9989858012170385, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 13.07, | |
| "learning_rate": 7.249860335195531e-05, | |
| "loss": 5.0055, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 13.57, | |
| "learning_rate": 7.207960893854748e-05, | |
| "loss": 4.8755, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 7.166480446927373e-05, | |
| "loss": 4.7729, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 14.57, | |
| "learning_rate": 7.124581005586592e-05, | |
| "loss": 4.7114, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 15.08, | |
| "learning_rate": 7.083100558659217e-05, | |
| "loss": 4.6184, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 15.08, | |
| "eval_cer": 0.6767552995956377, | |
| "eval_loss": 3.83825421333313, | |
| "eval_runtime": 124.4317, | |
| "eval_samples_per_second": 23.772, | |
| "eval_steps_per_second": 2.974, | |
| "eval_wer": 0.9983096686950642, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 15.58, | |
| "learning_rate": 7.041201117318434e-05, | |
| "loss": 4.5657, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 16.08, | |
| "learning_rate": 6.999301675977653e-05, | |
| "loss": 4.5613, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 16.58, | |
| "learning_rate": 6.95782122905028e-05, | |
| "loss": 4.4679, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 17.09, | |
| "learning_rate": 6.915921787709497e-05, | |
| "loss": 4.4288, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "learning_rate": 6.874022346368715e-05, | |
| "loss": 4.365, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "eval_cer": 0.62994731037863, | |
| "eval_loss": 3.4632537364959717, | |
| "eval_runtime": 122.6998, | |
| "eval_samples_per_second": 24.108, | |
| "eval_steps_per_second": 3.015, | |
| "eval_wer": 0.9959432048681541, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 18.09, | |
| "learning_rate": 6.832122905027933e-05, | |
| "loss": 4.3607, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 18.59, | |
| "learning_rate": 6.790223463687151e-05, | |
| "loss": 4.2659, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 19.1, | |
| "learning_rate": 6.748324022346368e-05, | |
| "loss": 4.2805, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 19.6, | |
| "learning_rate": 6.706424581005587e-05, | |
| "loss": 4.1783, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 20.1, | |
| "learning_rate": 6.664525139664804e-05, | |
| "loss": 4.1026, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 20.1, | |
| "eval_cer": 0.5813830004492914, | |
| "eval_loss": 3.073154926300049, | |
| "eval_runtime": 122.4373, | |
| "eval_samples_per_second": 24.159, | |
| "eval_steps_per_second": 3.022, | |
| "eval_wer": 0.9901960784313726, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 20.6, | |
| "learning_rate": 6.622625698324022e-05, | |
| "loss": 4.0481, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 21.11, | |
| "learning_rate": 6.58072625698324e-05, | |
| "loss": 4.005, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 21.61, | |
| "learning_rate": 6.538826815642457e-05, | |
| "loss": 3.966, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 22.11, | |
| "learning_rate": 6.496927374301676e-05, | |
| "loss": 3.9327, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 22.61, | |
| "learning_rate": 6.455027932960893e-05, | |
| "loss": 3.8655, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 22.61, | |
| "eval_cer": 0.5465016542090431, | |
| "eval_loss": 2.7638278007507324, | |
| "eval_runtime": 121.1871, | |
| "eval_samples_per_second": 24.409, | |
| "eval_steps_per_second": 3.053, | |
| "eval_wer": 0.986815415821501, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 23.12, | |
| "learning_rate": 6.413128491620112e-05, | |
| "loss": 3.8466, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 23.62, | |
| "learning_rate": 6.371229050279329e-05, | |
| "loss": 3.7785, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 24.12, | |
| "learning_rate": 6.329329608938548e-05, | |
| "loss": 3.7262, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 24.62, | |
| "learning_rate": 6.287430167597765e-05, | |
| "loss": 3.6768, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 25.13, | |
| "learning_rate": 6.245530726256982e-05, | |
| "loss": 3.6991, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 25.13, | |
| "eval_cer": 0.5088020258955194, | |
| "eval_loss": 2.475937604904175, | |
| "eval_runtime": 122.6664, | |
| "eval_samples_per_second": 24.114, | |
| "eval_steps_per_second": 3.016, | |
| "eval_wer": 0.9810682893847193, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 25.63, | |
| "learning_rate": 6.203631284916201e-05, | |
| "loss": 3.5971, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 26.13, | |
| "learning_rate": 6.161731843575418e-05, | |
| "loss": 3.6243, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 26.63, | |
| "learning_rate": 6.119832402234637e-05, | |
| "loss": 3.5803, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 27.14, | |
| "learning_rate": 6.077932960893854e-05, | |
| "loss": 3.5697, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 27.64, | |
| "learning_rate": 6.03645251396648e-05, | |
| "loss": 3.4894, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 27.64, | |
| "eval_cer": 0.4851529632806437, | |
| "eval_loss": 2.2937276363372803, | |
| "eval_runtime": 122.0039, | |
| "eval_samples_per_second": 24.245, | |
| "eval_steps_per_second": 3.033, | |
| "eval_wer": 0.9746450304259635, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 28.14, | |
| "learning_rate": 5.994553072625698e-05, | |
| "loss": 3.5363, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 28.64, | |
| "learning_rate": 5.952653631284916e-05, | |
| "loss": 3.4932, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 29.15, | |
| "learning_rate": 5.910754189944134e-05, | |
| "loss": 3.436, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 29.65, | |
| "learning_rate": 5.868854748603351e-05, | |
| "loss": 3.4209, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 30.15, | |
| "learning_rate": 5.8269553072625696e-05, | |
| "loss": 3.3983, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 30.15, | |
| "eval_cer": 0.4673855328186905, | |
| "eval_loss": 2.168361186981201, | |
| "eval_runtime": 121.2096, | |
| "eval_samples_per_second": 24.404, | |
| "eval_steps_per_second": 3.053, | |
| "eval_wer": 0.9732927653820149, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 30.65, | |
| "learning_rate": 5.785055865921787e-05, | |
| "loss": 3.3573, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 31.16, | |
| "learning_rate": 5.743156424581005e-05, | |
| "loss": 3.3821, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 31.66, | |
| "learning_rate": 5.701256983240223e-05, | |
| "loss": 3.3188, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 32.16, | |
| "learning_rate": 5.659357541899441e-05, | |
| "loss": 3.3519, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 32.66, | |
| "learning_rate": 5.6174581005586586e-05, | |
| "loss": 3.2736, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 32.66, | |
| "eval_cer": 0.44581954825797493, | |
| "eval_loss": 2.037248134613037, | |
| "eval_runtime": 123.5198, | |
| "eval_samples_per_second": 23.948, | |
| "eval_steps_per_second": 2.995, | |
| "eval_wer": 0.9658553076402975, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 33.17, | |
| "learning_rate": 5.5755586592178765e-05, | |
| "loss": 3.3093, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 33.67, | |
| "learning_rate": 5.5336592178770945e-05, | |
| "loss": 3.2599, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 34.17, | |
| "learning_rate": 5.4917597765363124e-05, | |
| "loss": 3.2115, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 34.67, | |
| "learning_rate": 5.4498603351955304e-05, | |
| "loss": 3.1972, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 35.18, | |
| "learning_rate": 5.407960893854748e-05, | |
| "loss": 3.1884, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 35.18, | |
| "eval_cer": 0.4329126332557285, | |
| "eval_loss": 1.9266635179519653, | |
| "eval_runtime": 122.7815, | |
| "eval_samples_per_second": 24.092, | |
| "eval_steps_per_second": 3.013, | |
| "eval_wer": 0.964841108857336, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 35.68, | |
| "learning_rate": 5.366061452513966e-05, | |
| "loss": 3.1908, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 36.18, | |
| "learning_rate": 5.324162011173184e-05, | |
| "loss": 3.1505, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 36.68, | |
| "learning_rate": 5.2822625698324014e-05, | |
| "loss": 3.134, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 37.19, | |
| "learning_rate": 5.24036312849162e-05, | |
| "loss": 3.101, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 37.69, | |
| "learning_rate": 5.198882681564245e-05, | |
| "loss": 3.1248, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 37.69, | |
| "eval_cer": 0.4217211942980844, | |
| "eval_loss": 1.8408104181289673, | |
| "eval_runtime": 122.5199, | |
| "eval_samples_per_second": 24.143, | |
| "eval_steps_per_second": 3.02, | |
| "eval_wer": 0.9590939824205544, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 38.19, | |
| "learning_rate": 5.156983240223463e-05, | |
| "loss": 3.0958, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 38.69, | |
| "learning_rate": 5.115083798882681e-05, | |
| "loss": 3.0627, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 39.2, | |
| "learning_rate": 5.073184357541899e-05, | |
| "loss": 3.0716, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 39.7, | |
| "learning_rate": 5.031284916201117e-05, | |
| "loss": 3.0455, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 40.2, | |
| "learning_rate": 4.989804469273743e-05, | |
| "loss": 3.0381, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 40.2, | |
| "eval_cer": 0.40742556059306456, | |
| "eval_loss": 1.7530696392059326, | |
| "eval_runtime": 125.5769, | |
| "eval_samples_per_second": 23.555, | |
| "eval_steps_per_second": 2.946, | |
| "eval_wer": 0.9503042596348884, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 40.7, | |
| "learning_rate": 4.94790502793296e-05, | |
| "loss": 2.9917, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 41.21, | |
| "learning_rate": 4.906005586592179e-05, | |
| "loss": 3.0649, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 41.71, | |
| "learning_rate": 4.864106145251396e-05, | |
| "loss": 2.9746, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 42.21, | |
| "learning_rate": 4.8222067039106146e-05, | |
| "loss": 2.9459, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 42.71, | |
| "learning_rate": 4.780307262569832e-05, | |
| "loss": 2.9515, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 42.71, | |
| "eval_cer": 0.39672425764816405, | |
| "eval_loss": 1.6879578828811646, | |
| "eval_runtime": 121.7673, | |
| "eval_samples_per_second": 24.292, | |
| "eval_steps_per_second": 3.039, | |
| "eval_wer": 0.9459093982420554, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 43.22, | |
| "learning_rate": 4.7384078212290505e-05, | |
| "loss": 2.9478, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 43.72, | |
| "learning_rate": 4.696508379888268e-05, | |
| "loss": 2.9046, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 44.22, | |
| "learning_rate": 4.654608938547485e-05, | |
| "loss": 2.9345, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 44.72, | |
| "learning_rate": 4.6127094972067036e-05, | |
| "loss": 2.8722, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 45.23, | |
| "learning_rate": 4.570810055865921e-05, | |
| "loss": 2.8704, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 45.23, | |
| "eval_cer": 0.3884327900992525, | |
| "eval_loss": 1.626428484916687, | |
| "eval_runtime": 124.9171, | |
| "eval_samples_per_second": 23.68, | |
| "eval_steps_per_second": 2.962, | |
| "eval_wer": 0.9377958079783637, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 45.73, | |
| "learning_rate": 4.5289106145251395e-05, | |
| "loss": 2.8807, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 46.23, | |
| "learning_rate": 4.487011173184357e-05, | |
| "loss": 2.8631, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 46.73, | |
| "learning_rate": 4.4451117318435753e-05, | |
| "loss": 2.8743, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 47.24, | |
| "learning_rate": 4.4032122905027926e-05, | |
| "loss": 2.8364, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 47.74, | |
| "learning_rate": 4.361312849162011e-05, | |
| "loss": 2.8128, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 47.74, | |
| "eval_cer": 0.37818077849936693, | |
| "eval_loss": 1.5620697736740112, | |
| "eval_runtime": 128.811, | |
| "eval_samples_per_second": 22.964, | |
| "eval_steps_per_second": 2.872, | |
| "eval_wer": 0.934077079107505, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 48.24, | |
| "learning_rate": 4.3194134078212285e-05, | |
| "loss": 2.855, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 48.74, | |
| "learning_rate": 4.2775139664804464e-05, | |
| "loss": 2.7587, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 49.25, | |
| "learning_rate": 4.2356145251396644e-05, | |
| "loss": 2.7794, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 49.75, | |
| "learning_rate": 4.193715083798882e-05, | |
| "loss": 2.7664, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 50.25, | |
| "learning_rate": 4.1518156424581e-05, | |
| "loss": 2.7386, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 50.25, | |
| "eval_cer": 0.3663766695257934, | |
| "eval_loss": 1.5010998249053955, | |
| "eval_runtime": 124.8062, | |
| "eval_samples_per_second": 23.701, | |
| "eval_steps_per_second": 2.965, | |
| "eval_wer": 0.9242731575388776, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 50.75, | |
| "learning_rate": 4.109916201117318e-05, | |
| "loss": 2.7295, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 51.26, | |
| "learning_rate": 4.0684357541899433e-05, | |
| "loss": 2.7372, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 51.76, | |
| "learning_rate": 4.026536312849162e-05, | |
| "loss": 2.7116, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 52.26, | |
| "learning_rate": 3.984636871508379e-05, | |
| "loss": 2.6882, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 52.76, | |
| "learning_rate": 3.942737430167598e-05, | |
| "loss": 2.6646, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 52.76, | |
| "eval_cer": 0.3574725319609525, | |
| "eval_loss": 1.460774540901184, | |
| "eval_runtime": 125.7182, | |
| "eval_samples_per_second": 23.529, | |
| "eval_steps_per_second": 2.943, | |
| "eval_wer": 0.9192021636240704, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 53.27, | |
| "learning_rate": 3.900837988826815e-05, | |
| "loss": 2.6904, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 53.77, | |
| "learning_rate": 3.858938547486034e-05, | |
| "loss": 2.6589, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 54.27, | |
| "learning_rate": 3.817039106145251e-05, | |
| "loss": 2.6618, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 54.77, | |
| "learning_rate": 3.7751396648044696e-05, | |
| "loss": 2.6187, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 55.28, | |
| "learning_rate": 3.733240223463687e-05, | |
| "loss": 2.6072, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 55.28, | |
| "eval_cer": 0.3501204917697995, | |
| "eval_loss": 1.4251333475112915, | |
| "eval_runtime": 122.3324, | |
| "eval_samples_per_second": 24.18, | |
| "eval_steps_per_second": 3.025, | |
| "eval_wer": 0.9148073022312373, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 55.78, | |
| "learning_rate": 3.691340782122905e-05, | |
| "loss": 2.6307, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 56.28, | |
| "learning_rate": 3.649441340782123e-05, | |
| "loss": 2.5884, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 56.78, | |
| "learning_rate": 3.6075418994413407e-05, | |
| "loss": 2.5678, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 57.29, | |
| "learning_rate": 3.5656424581005586e-05, | |
| "loss": 2.5722, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 57.79, | |
| "learning_rate": 3.523743016759776e-05, | |
| "loss": 2.569, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 57.79, | |
| "eval_cer": 0.34615855900012255, | |
| "eval_loss": 1.3836983442306519, | |
| "eval_runtime": 125.4562, | |
| "eval_samples_per_second": 23.578, | |
| "eval_steps_per_second": 2.949, | |
| "eval_wer": 0.9060175794455714, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 58.29, | |
| "learning_rate": 3.481843575418994e-05, | |
| "loss": 2.5487, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 58.79, | |
| "learning_rate": 3.439944134078212e-05, | |
| "loss": 2.5536, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 59.3, | |
| "learning_rate": 3.39804469273743e-05, | |
| "loss": 2.5402, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 59.8, | |
| "learning_rate": 3.3561452513966476e-05, | |
| "loss": 2.5001, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 60.3, | |
| "learning_rate": 3.3142458100558655e-05, | |
| "loss": 2.5091, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 60.3, | |
| "eval_cer": 0.3392149654862558, | |
| "eval_loss": 1.3589394092559814, | |
| "eval_runtime": 123.5116, | |
| "eval_samples_per_second": 23.949, | |
| "eval_steps_per_second": 2.996, | |
| "eval_wer": 0.9070317782285328, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 60.8, | |
| "learning_rate": 3.2723463687150835e-05, | |
| "loss": 2.4874, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 61.31, | |
| "learning_rate": 3.2304469273743014e-05, | |
| "loss": 2.4834, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 61.81, | |
| "learning_rate": 3.1885474860335194e-05, | |
| "loss": 2.478, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 62.31, | |
| "learning_rate": 3.147067039106145e-05, | |
| "loss": 2.4811, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 62.81, | |
| "learning_rate": 3.105167597765363e-05, | |
| "loss": 2.4588, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 62.81, | |
| "eval_cer": 0.3283911285381693, | |
| "eval_loss": 1.326094627380371, | |
| "eval_runtime": 126.176, | |
| "eval_samples_per_second": 23.443, | |
| "eval_steps_per_second": 2.932, | |
| "eval_wer": 0.896551724137931, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 63.32, | |
| "learning_rate": 3.063268156424581e-05, | |
| "loss": 2.4296, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 63.82, | |
| "learning_rate": 3.0213687150837987e-05, | |
| "loss": 2.4535, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 64.32, | |
| "learning_rate": 2.9794692737430163e-05, | |
| "loss": 2.4572, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 64.82, | |
| "learning_rate": 2.9375698324022342e-05, | |
| "loss": 2.3824, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 65.33, | |
| "learning_rate": 2.896508379888268e-05, | |
| "loss": 2.4083, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 65.33, | |
| "eval_cer": 0.32647142915492383, | |
| "eval_loss": 1.3052246570587158, | |
| "eval_runtime": 125.7147, | |
| "eval_samples_per_second": 23.529, | |
| "eval_steps_per_second": 2.943, | |
| "eval_wer": 0.8982420554428668, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 65.83, | |
| "learning_rate": 2.854608938547486e-05, | |
| "loss": 2.4015, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 66.33, | |
| "learning_rate": 2.812709497206704e-05, | |
| "loss": 2.3747, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 66.83, | |
| "learning_rate": 2.7708100558659218e-05, | |
| "loss": 2.3625, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 67.34, | |
| "learning_rate": 2.728910614525139e-05, | |
| "loss": 2.4027, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 67.84, | |
| "learning_rate": 2.687011173184357e-05, | |
| "loss": 2.3787, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 67.84, | |
| "eval_cer": 0.32430666176530654, | |
| "eval_loss": 1.2997361421585083, | |
| "eval_runtime": 122.085, | |
| "eval_samples_per_second": 24.229, | |
| "eval_steps_per_second": 3.031, | |
| "eval_wer": 0.8908045977011494, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 68.34, | |
| "learning_rate": 2.645111731843575e-05, | |
| "loss": 2.3351, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 68.84, | |
| "learning_rate": 2.603212290502793e-05, | |
| "loss": 2.3445, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 69.35, | |
| "learning_rate": 2.561312849162011e-05, | |
| "loss": 2.3649, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 69.85, | |
| "learning_rate": 2.5194134078212288e-05, | |
| "loss": 2.326, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 70.35, | |
| "learning_rate": 2.4775139664804467e-05, | |
| "loss": 2.3457, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 70.35, | |
| "eval_cer": 0.3187109422864845, | |
| "eval_loss": 1.2778037786483765, | |
| "eval_runtime": 128.0452, | |
| "eval_samples_per_second": 23.101, | |
| "eval_steps_per_second": 2.89, | |
| "eval_wer": 0.889790398918188, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 70.85, | |
| "learning_rate": 2.4356145251396646e-05, | |
| "loss": 2.2999, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 71.36, | |
| "learning_rate": 2.3937150837988826e-05, | |
| "loss": 2.3383, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 71.86, | |
| "learning_rate": 2.3518156424581005e-05, | |
| "loss": 2.312, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 72.36, | |
| "learning_rate": 2.309916201117318e-05, | |
| "loss": 2.296, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 72.86, | |
| "learning_rate": 2.268016759776536e-05, | |
| "loss": 2.3099, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 72.86, | |
| "eval_cer": 0.3172405342482539, | |
| "eval_loss": 1.266068935394287, | |
| "eval_runtime": 123.4423, | |
| "eval_samples_per_second": 23.963, | |
| "eval_steps_per_second": 2.997, | |
| "eval_wer": 0.8830290736984449, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 73.37, | |
| "learning_rate": 2.226117318435754e-05, | |
| "loss": 2.2976, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 73.87, | |
| "learning_rate": 2.184217877094972e-05, | |
| "loss": 2.3002, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 74.37, | |
| "learning_rate": 2.14231843575419e-05, | |
| "loss": 2.2536, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 74.87, | |
| "learning_rate": 2.1004189944134078e-05, | |
| "loss": 2.2683, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 75.38, | |
| "learning_rate": 2.0585195530726257e-05, | |
| "loss": 2.2559, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 75.38, | |
| "eval_cer": 0.3143405628395213, | |
| "eval_loss": 1.2474771738052368, | |
| "eval_runtime": 124.7835, | |
| "eval_samples_per_second": 23.705, | |
| "eval_steps_per_second": 2.965, | |
| "eval_wer": 0.8850574712643678, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 75.88, | |
| "learning_rate": 2.0166201117318437e-05, | |
| "loss": 2.2334, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 76.38, | |
| "learning_rate": 1.9747206703910616e-05, | |
| "loss": 2.2153, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 76.88, | |
| "learning_rate": 1.932821229050279e-05, | |
| "loss": 2.2244, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 77.39, | |
| "learning_rate": 1.8909217877094968e-05, | |
| "loss": 2.2327, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 77.89, | |
| "learning_rate": 1.8490223463687148e-05, | |
| "loss": 2.2264, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 77.89, | |
| "eval_cer": 0.3085406200220561, | |
| "eval_loss": 1.2318875789642334, | |
| "eval_runtime": 127.717, | |
| "eval_samples_per_second": 23.161, | |
| "eval_steps_per_second": 2.897, | |
| "eval_wer": 0.8739012846517917, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 78.39, | |
| "learning_rate": 1.8071229050279327e-05, | |
| "loss": 2.2722, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 78.89, | |
| "learning_rate": 1.7652234636871506e-05, | |
| "loss": 2.1997, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 79.4, | |
| "learning_rate": 1.7233240223463686e-05, | |
| "loss": 2.1979, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 79.9, | |
| "learning_rate": 1.6814245810055865e-05, | |
| "loss": 2.222, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 80.4, | |
| "learning_rate": 1.639525139664804e-05, | |
| "loss": 2.196, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 80.4, | |
| "eval_cer": 0.3048645999264796, | |
| "eval_loss": 1.2218027114868164, | |
| "eval_runtime": 124.9046, | |
| "eval_samples_per_second": 23.682, | |
| "eval_steps_per_second": 2.962, | |
| "eval_wer": 0.8722109533468559, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 80.9, | |
| "learning_rate": 1.597625698324022e-05, | |
| "loss": 2.2138, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 81.41, | |
| "learning_rate": 1.55572625698324e-05, | |
| "loss": 2.1853, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 81.91, | |
| "learning_rate": 1.5138268156424581e-05, | |
| "loss": 2.2029, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 82.41, | |
| "learning_rate": 1.471927374301676e-05, | |
| "loss": 2.1583, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 82.91, | |
| "learning_rate": 1.4300279329608936e-05, | |
| "loss": 2.1613, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 82.91, | |
| "eval_cer": 0.30506882326512275, | |
| "eval_loss": 1.2093260288238525, | |
| "eval_runtime": 123.6333, | |
| "eval_samples_per_second": 23.926, | |
| "eval_steps_per_second": 2.993, | |
| "eval_wer": 0.8718728870858689, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 83.42, | |
| "learning_rate": 1.3881284916201116e-05, | |
| "loss": 2.1637, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 83.92, | |
| "learning_rate": 1.3462290502793295e-05, | |
| "loss": 2.1648, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 84.42, | |
| "learning_rate": 1.3043296089385474e-05, | |
| "loss": 2.1639, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 84.92, | |
| "learning_rate": 1.2624301675977652e-05, | |
| "loss": 2.1513, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 85.43, | |
| "learning_rate": 1.2205307262569831e-05, | |
| "loss": 2.1455, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 85.43, | |
| "eval_cer": 0.30053506514724504, | |
| "eval_loss": 1.2055062055587769, | |
| "eval_runtime": 124.9017, | |
| "eval_samples_per_second": 23.683, | |
| "eval_steps_per_second": 2.962, | |
| "eval_wer": 0.8624070317782285, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 85.93, | |
| "learning_rate": 1.178631284916201e-05, | |
| "loss": 2.1356, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 86.43, | |
| "learning_rate": 1.1367318435754188e-05, | |
| "loss": 2.1369, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 86.93, | |
| "learning_rate": 1.0948324022346368e-05, | |
| "loss": 2.1328, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 87.44, | |
| "learning_rate": 1.0529329608938546e-05, | |
| "loss": 2.1225, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 87.94, | |
| "learning_rate": 1.0110335195530725e-05, | |
| "loss": 2.1193, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 87.94, | |
| "eval_cer": 0.29824776375444184, | |
| "eval_loss": 1.1974669694900513, | |
| "eval_runtime": 123.9455, | |
| "eval_samples_per_second": 23.865, | |
| "eval_steps_per_second": 2.985, | |
| "eval_wer": 0.8600405679513184, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 88.44, | |
| "learning_rate": 9.691340782122904e-06, | |
| "loss": 2.1388, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 88.94, | |
| "learning_rate": 9.276536312849161e-06, | |
| "loss": 2.0962, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 89.45, | |
| "learning_rate": 8.85754189944134e-06, | |
| "loss": 2.1021, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 89.95, | |
| "learning_rate": 8.438547486033518e-06, | |
| "loss": 2.0829, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 90.45, | |
| "learning_rate": 8.019553072625698e-06, | |
| "loss": 2.0911, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 90.45, | |
| "eval_cer": 0.30028999714087323, | |
| "eval_loss": 1.1960209608078003, | |
| "eval_runtime": 122.0182, | |
| "eval_samples_per_second": 24.242, | |
| "eval_steps_per_second": 3.032, | |
| "eval_wer": 0.8647734956051386, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 90.95, | |
| "learning_rate": 7.600558659217876e-06, | |
| "loss": 2.0836, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 91.46, | |
| "learning_rate": 7.1815642458100555e-06, | |
| "loss": 2.0527, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 91.96, | |
| "learning_rate": 6.762569832402233e-06, | |
| "loss": 2.0807, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 92.46, | |
| "learning_rate": 6.3435754189944126e-06, | |
| "loss": 2.0962, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 92.96, | |
| "learning_rate": 5.924581005586592e-06, | |
| "loss": 2.0884, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 92.96, | |
| "eval_cer": 0.2971449577257689, | |
| "eval_loss": 1.1871271133422852, | |
| "eval_runtime": 124.4024, | |
| "eval_samples_per_second": 23.778, | |
| "eval_steps_per_second": 2.974, | |
| "eval_wer": 0.8637592968221771, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 93.47, | |
| "learning_rate": 5.5055865921787705e-06, | |
| "loss": 2.0708, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 93.97, | |
| "learning_rate": 5.08659217877095e-06, | |
| "loss": 2.067, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 94.47, | |
| "learning_rate": 4.667597765363128e-06, | |
| "loss": 2.0985, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 94.97, | |
| "learning_rate": 4.248603351955307e-06, | |
| "loss": 2.0655, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 95.48, | |
| "learning_rate": 3.829608938547485e-06, | |
| "loss": 2.0766, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 95.48, | |
| "eval_cer": 0.296695666380754, | |
| "eval_loss": 1.1813979148864746, | |
| "eval_runtime": 123.6675, | |
| "eval_samples_per_second": 23.919, | |
| "eval_steps_per_second": 2.992, | |
| "eval_wer": 0.8617308992562542, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 95.98, | |
| "learning_rate": 3.4106145251396644e-06, | |
| "loss": 2.0485, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 96.48, | |
| "learning_rate": 2.9916201117318433e-06, | |
| "loss": 2.0725, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 96.98, | |
| "learning_rate": 2.572625698324022e-06, | |
| "loss": 2.047, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 97.49, | |
| "learning_rate": 2.153631284916201e-06, | |
| "loss": 2.0649, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 97.99, | |
| "learning_rate": 1.7346368715083795e-06, | |
| "loss": 2.0735, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 97.99, | |
| "eval_cer": 0.2968590450516685, | |
| "eval_loss": 1.180108904838562, | |
| "eval_runtime": 123.9462, | |
| "eval_samples_per_second": 23.865, | |
| "eval_steps_per_second": 2.985, | |
| "eval_wer": 0.8620689655172413, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 98.49, | |
| "learning_rate": 1.3156424581005587e-06, | |
| "loss": 2.0277, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 98.99, | |
| "learning_rate": 8.966480446927373e-07, | |
| "loss": 2.0545, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 99.5, | |
| "learning_rate": 4.776536312849162e-07, | |
| "loss": 2.0502, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 5.865921787709496e-08, | |
| "loss": 2.0333, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 19900, | |
| "total_flos": 6.621453767453566e+19, | |
| "train_loss": 5.91964619928868, | |
| "train_runtime": 36587.1788, | |
| "train_samples_per_second": 17.438, | |
| "train_steps_per_second": 0.544 | |
| } | |
| ], | |
| "max_steps": 19900, | |
| "num_train_epochs": 100, | |
| "total_flos": 6.621453767453566e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |