| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 14.999677731227845, | |
| "global_step": 23265, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 5.4592, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9.800000000000001e-06, | |
| "loss": 2.9124, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.48e-05, | |
| "loss": 2.3526, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 1.9800000000000004e-05, | |
| "loss": 1.1889, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 2.48e-05, | |
| "loss": 0.8361, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": 0.6304140686988831, | |
| "eval_runtime": 553.4275, | |
| "eval_samples_per_second": 9.824, | |
| "eval_steps_per_second": 0.614, | |
| "eval_wer": 0.4970241305264396, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 2.98e-05, | |
| "loss": 0.6758, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.48e-05, | |
| "loss": 0.6902, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 3.9800000000000005e-05, | |
| "loss": 0.6335, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.4800000000000005e-05, | |
| "loss": 0.5867, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.9800000000000004e-05, | |
| "loss": 0.5703, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 0.3195364475250244, | |
| "eval_runtime": 552.5688, | |
| "eval_samples_per_second": 9.839, | |
| "eval_steps_per_second": 0.615, | |
| "eval_wer": 0.27750595664153277, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 5.4800000000000004e-05, | |
| "loss": 0.5762, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 5.9800000000000003e-05, | |
| "loss": 0.5656, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 6.48e-05, | |
| "loss": 0.5864, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 6.975e-05, | |
| "loss": 0.5116, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 7.475000000000001e-05, | |
| "loss": 0.5451, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_loss": 0.2699669599533081, | |
| "eval_runtime": 552.0028, | |
| "eval_samples_per_second": 9.85, | |
| "eval_steps_per_second": 0.616, | |
| "eval_wer": 0.2246070577622638, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 7.975e-05, | |
| "loss": 0.5134, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 8.475000000000001e-05, | |
| "loss": 0.467, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 8.975e-05, | |
| "loss": 0.4769, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 9.47e-05, | |
| "loss": 0.461, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 9.970000000000001e-05, | |
| "loss": 0.47, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_loss": 0.25637131929397583, | |
| "eval_runtime": 552.4533, | |
| "eval_samples_per_second": 9.842, | |
| "eval_steps_per_second": 0.615, | |
| "eval_wer": 0.23294145332248228, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 9.955795908770281e-05, | |
| "loss": 0.4506, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 9.908770279802493e-05, | |
| "loss": 0.4154, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 9.861744650834705e-05, | |
| "loss": 0.4377, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 9.814719021866918e-05, | |
| "loss": 0.4723, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 9.767693392899132e-05, | |
| "loss": 0.4063, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "eval_loss": 0.2459278106689453, | |
| "eval_runtime": 552.824, | |
| "eval_samples_per_second": 9.835, | |
| "eval_steps_per_second": 0.615, | |
| "eval_wer": 0.20988949572004276, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 9.720667763931343e-05, | |
| "loss": 0.3807, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 9.673642134963555e-05, | |
| "loss": 0.4094, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 9.626616505995768e-05, | |
| "loss": 0.401, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 9.579590877027981e-05, | |
| "loss": 0.3849, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 9.532565248060194e-05, | |
| "loss": 0.374, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "eval_loss": 0.21745240688323975, | |
| "eval_runtime": 553.0877, | |
| "eval_samples_per_second": 9.83, | |
| "eval_steps_per_second": 0.615, | |
| "eval_wer": 0.18940649298439996, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.485539619092406e-05, | |
| "loss": 0.3684, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 9.438513990124619e-05, | |
| "loss": 0.3658, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 9.391488361156831e-05, | |
| "loss": 0.3432, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 9.344462732189043e-05, | |
| "loss": 0.336, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 9.297437103221256e-05, | |
| "loss": 0.3297, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 0.20356999337673187, | |
| "eval_runtime": 553.8668, | |
| "eval_samples_per_second": 9.816, | |
| "eval_steps_per_second": 0.614, | |
| "eval_wer": 0.17550276015570612, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 9.250411474253468e-05, | |
| "loss": 0.3177, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 9.203385845285682e-05, | |
| "loss": 0.3111, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 9.156360216317894e-05, | |
| "loss": 0.3209, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 9.109334587350105e-05, | |
| "loss": 0.3439, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 9.062308958382318e-05, | |
| "loss": 0.3145, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "eval_loss": 0.1957111805677414, | |
| "eval_runtime": 553.0089, | |
| "eval_samples_per_second": 9.832, | |
| "eval_steps_per_second": 0.615, | |
| "eval_wer": 0.1757282790943944, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 9.01575358570421e-05, | |
| "loss": 0.3077, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 8.968727956736421e-05, | |
| "loss": 0.3147, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 8.921702327768633e-05, | |
| "loss": 0.3232, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 8.874676698800847e-05, | |
| "loss": 0.3195, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 8.82765106983306e-05, | |
| "loss": 0.3989, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_loss": 0.19227643311023712, | |
| "eval_runtime": 554.5698, | |
| "eval_samples_per_second": 9.804, | |
| "eval_steps_per_second": 0.613, | |
| "eval_wer": 0.17229646915783384, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 8.780625440865272e-05, | |
| "loss": 0.3035, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 8.733599811897484e-05, | |
| "loss": 0.2997, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 8.686574182929698e-05, | |
| "loss": 0.2798, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 8.639548553961909e-05, | |
| "loss": 0.274, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 8.592522924994122e-05, | |
| "loss": 0.271, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "eval_loss": 0.18889272212982178, | |
| "eval_runtime": 553.6667, | |
| "eval_samples_per_second": 9.82, | |
| "eval_steps_per_second": 0.614, | |
| "eval_wer": 0.16491317520860502, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 8.545497296026334e-05, | |
| "loss": 0.2726, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 8.498471667058548e-05, | |
| "loss": 0.283, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 8.45144603809076e-05, | |
| "loss": 0.2837, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 8.404420409122973e-05, | |
| "loss": 0.2915, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 8.357394780155185e-05, | |
| "loss": 0.2758, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "eval_loss": 0.17681501805782318, | |
| "eval_runtime": 557.2307, | |
| "eval_samples_per_second": 9.757, | |
| "eval_steps_per_second": 0.61, | |
| "eval_wer": 0.15884377420651652, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 8.310369151187397e-05, | |
| "loss": 0.2699, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 8.26334352221961e-05, | |
| "loss": 0.2659, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 8.216317893251822e-05, | |
| "loss": 0.2759, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 8.169292264284035e-05, | |
| "loss": 0.2632, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 8.122736891605925e-05, | |
| "loss": 0.2683, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "eval_loss": 0.17198732495307922, | |
| "eval_runtime": 555.6592, | |
| "eval_samples_per_second": 9.785, | |
| "eval_steps_per_second": 0.612, | |
| "eval_wer": 0.1533822938217616, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 8.075711262638138e-05, | |
| "loss": 0.2628, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 8.02868563367035e-05, | |
| "loss": 0.3404, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 7.981660004702564e-05, | |
| "loss": 0.2964, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 7.934634375734776e-05, | |
| "loss": 0.2458, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 7.887608746766987e-05, | |
| "loss": 0.2341, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "eval_loss": 0.16891594231128693, | |
| "eval_runtime": 553.369, | |
| "eval_samples_per_second": 9.825, | |
| "eval_steps_per_second": 0.614, | |
| "eval_wer": 0.1470775687097375, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 7.840583117799201e-05, | |
| "loss": 0.2353, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 7.794027745121092e-05, | |
| "loss": 0.2433, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 7.747002116153303e-05, | |
| "loss": 0.2467, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 7.699976487185517e-05, | |
| "loss": 0.2423, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 7.652950858217729e-05, | |
| "loss": 0.2316, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "eval_loss": 0.1705636978149414, | |
| "eval_runtime": 555.4457, | |
| "eval_samples_per_second": 9.789, | |
| "eval_steps_per_second": 0.612, | |
| "eval_wer": 0.14053751948777785, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 7.605925229249942e-05, | |
| "loss": 0.2239, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 7.558899600282154e-05, | |
| "loss": 0.2305, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 7.511873971314367e-05, | |
| "loss": 0.2291, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 7.464848342346579e-05, | |
| "loss": 0.2411, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 7.417822713378791e-05, | |
| "loss": 0.2383, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "eval_loss": 0.16366632282733917, | |
| "eval_runtime": 555.2596, | |
| "eval_samples_per_second": 9.792, | |
| "eval_steps_per_second": 0.612, | |
| "eval_wer": 0.1426064106209615, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 7.370797084411004e-05, | |
| "loss": 0.2328, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 7.323771455443218e-05, | |
| "loss": 0.2575, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 7.27674582647543e-05, | |
| "loss": 0.2183, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 7.229720197507642e-05, | |
| "loss": 0.2254, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 7.182694568539855e-05, | |
| "loss": 0.2148, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "eval_loss": 0.1583578884601593, | |
| "eval_runtime": 555.4598, | |
| "eval_samples_per_second": 9.788, | |
| "eval_steps_per_second": 0.612, | |
| "eval_wer": 0.1347328581093669, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 7.135668939572067e-05, | |
| "loss": 0.2096, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 7.08864331060428e-05, | |
| "loss": 0.2142, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 7.041617681636492e-05, | |
| "loss": 0.2113, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 5.42, | |
| "learning_rate": 6.994592052668704e-05, | |
| "loss": 0.2177, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 6.947566423700918e-05, | |
| "loss": 0.2085, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "eval_loss": 0.16014662384986877, | |
| "eval_runtime": 556.0462, | |
| "eval_samples_per_second": 9.778, | |
| "eval_steps_per_second": 0.611, | |
| "eval_wer": 0.13866473177954053, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 6.90054079473313e-05, | |
| "loss": 0.2112, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 6.853515165765343e-05, | |
| "loss": 0.2051, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 6.806489536797554e-05, | |
| "loss": 0.2214, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "learning_rate": 6.759463907829768e-05, | |
| "loss": 0.2155, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 6.71243827886198e-05, | |
| "loss": 0.2944, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "eval_loss": 0.1565837264060974, | |
| "eval_runtime": 556.6846, | |
| "eval_samples_per_second": 9.767, | |
| "eval_steps_per_second": 0.611, | |
| "eval_wer": 0.1293596242658378, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "learning_rate": 6.665412649894193e-05, | |
| "loss": 0.2048, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 6.618387020926405e-05, | |
| "loss": 0.2089, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 6.571831648248296e-05, | |
| "loss": 0.1928, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 6.524806019280508e-05, | |
| "loss": 0.2435, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 6.47778039031272e-05, | |
| "loss": 0.1944, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "eval_loss": 0.14938245713710785, | |
| "eval_runtime": 556.2024, | |
| "eval_samples_per_second": 9.775, | |
| "eval_steps_per_second": 0.611, | |
| "eval_wer": 0.12710443487895517, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 6.430754761344934e-05, | |
| "loss": 0.1952, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 6.383729132377145e-05, | |
| "loss": 0.1923, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "learning_rate": 6.336703503409358e-05, | |
| "loss": 0.1796, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "learning_rate": 6.28967787444157e-05, | |
| "loss": 0.1851, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 6.242652245473784e-05, | |
| "loss": 0.1853, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "eval_loss": 0.1561189442873001, | |
| "eval_runtime": 554.459, | |
| "eval_samples_per_second": 9.806, | |
| "eval_steps_per_second": 0.613, | |
| "eval_wer": 0.12469236275211547, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "learning_rate": 6.195626616505996e-05, | |
| "loss": 0.1918, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 6.148600987538209e-05, | |
| "loss": 0.1743, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "learning_rate": 6.101575358570422e-05, | |
| "loss": 0.1967, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 6.71, | |
| "learning_rate": 6.054549729602633e-05, | |
| "loss": 0.1922, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 6.0079943569245244e-05, | |
| "loss": 0.235, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "eval_loss": 0.14610491693019867, | |
| "eval_runtime": 555.3804, | |
| "eval_samples_per_second": 9.79, | |
| "eval_steps_per_second": 0.612, | |
| "eval_wer": 0.1214958769254905, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 5.9609687279567375e-05, | |
| "loss": 0.1934, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 5.9139430989889486e-05, | |
| "loss": 0.1913, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 5.866917470021162e-05, | |
| "loss": 0.1801, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 5.819891841053374e-05, | |
| "loss": 0.1829, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 5.772866212085587e-05, | |
| "loss": 0.2286, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "eval_loss": 0.1447422057390213, | |
| "eval_runtime": 557.5344, | |
| "eval_samples_per_second": 9.752, | |
| "eval_steps_per_second": 0.61, | |
| "eval_wer": 0.11670114818555306, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 7.16, | |
| "learning_rate": 5.7258405831177996e-05, | |
| "loss": 0.1764, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "learning_rate": 5.6788149541500127e-05, | |
| "loss": 0.1775, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 7.29, | |
| "learning_rate": 5.631789325182225e-05, | |
| "loss": 0.1602, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "learning_rate": 5.584763696214437e-05, | |
| "loss": 0.1748, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "learning_rate": 5.537738067246649e-05, | |
| "loss": 0.1781, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "eval_loss": 0.15018154680728912, | |
| "eval_runtime": 554.8903, | |
| "eval_samples_per_second": 9.798, | |
| "eval_steps_per_second": 0.613, | |
| "eval_wer": 0.11993685469716729, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 5.490712438278862e-05, | |
| "loss": 0.1682, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "learning_rate": 5.443686809311075e-05, | |
| "loss": 0.1706, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 7.61, | |
| "learning_rate": 5.397131436632965e-05, | |
| "loss": 0.1554, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 5.350105807665178e-05, | |
| "loss": 0.1612, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 5.30308017869739e-05, | |
| "loss": 0.1714, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "eval_loss": 0.14246602356433868, | |
| "eval_runtime": 557.0268, | |
| "eval_samples_per_second": 9.761, | |
| "eval_steps_per_second": 0.61, | |
| "eval_wer": 0.11787776873523097, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 5.2560545497296034e-05, | |
| "loss": 0.1719, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 5.209028920761816e-05, | |
| "loss": 0.1663, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 7.93, | |
| "learning_rate": 5.1620032917940275e-05, | |
| "loss": 0.1597, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "learning_rate": 5.11497766282624e-05, | |
| "loss": 0.1622, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 5.067952033858453e-05, | |
| "loss": 0.1725, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "eval_loss": 0.1426704078912735, | |
| "eval_runtime": 554.3005, | |
| "eval_samples_per_second": 9.809, | |
| "eval_steps_per_second": 0.613, | |
| "eval_wer": 0.11733848431662859, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 5.0209264048906654e-05, | |
| "loss": 0.1545, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 8.19, | |
| "learning_rate": 4.9739007759228785e-05, | |
| "loss": 0.1531, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "learning_rate": 4.92687514695509e-05, | |
| "loss": 0.1546, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "learning_rate": 4.8798495179873033e-05, | |
| "loss": 0.151, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "learning_rate": 4.832823889019516e-05, | |
| "loss": 0.143, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "eval_loss": 0.1447545289993286, | |
| "eval_runtime": 556.1274, | |
| "eval_samples_per_second": 9.777, | |
| "eval_steps_per_second": 0.611, | |
| "eval_wer": 0.11421063468873484, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 8.45, | |
| "learning_rate": 4.785798260051728e-05, | |
| "loss": 0.1568, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "learning_rate": 4.7387726310839406e-05, | |
| "loss": 0.1565, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 8.57, | |
| "learning_rate": 4.6917470021161537e-05, | |
| "loss": 0.1515, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "learning_rate": 4.644721373148366e-05, | |
| "loss": 0.2059, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 4.5976957441805785e-05, | |
| "loss": 0.154, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "eval_loss": 0.13924145698547363, | |
| "eval_runtime": 554.6978, | |
| "eval_samples_per_second": 9.802, | |
| "eval_steps_per_second": 0.613, | |
| "eval_wer": 0.11040622824477629, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 4.550670115212791e-05, | |
| "loss": 0.1606, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 8.83, | |
| "learning_rate": 4.504114742534681e-05, | |
| "loss": 0.1532, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "learning_rate": 4.457089113566894e-05, | |
| "loss": 0.1509, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "learning_rate": 4.4100634845991065e-05, | |
| "loss": 0.1538, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 4.3630378556313196e-05, | |
| "loss": 0.1447, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "eval_loss": 0.1403558999300003, | |
| "eval_runtime": 554.9504, | |
| "eval_samples_per_second": 9.797, | |
| "eval_steps_per_second": 0.613, | |
| "eval_wer": 0.1094159059487974, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 4.316012226663532e-05, | |
| "loss": 0.1445, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "learning_rate": 4.2689865976957444e-05, | |
| "loss": 0.1641, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "learning_rate": 4.221960968727957e-05, | |
| "loss": 0.1448, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 9.28, | |
| "learning_rate": 4.174935339760169e-05, | |
| "loss": 0.1386, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "learning_rate": 4.127909710792382e-05, | |
| "loss": 0.1471, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "eval_loss": 0.14039234817028046, | |
| "eval_runtime": 559.6371, | |
| "eval_samples_per_second": 9.715, | |
| "eval_steps_per_second": 0.608, | |
| "eval_wer": 0.10880798533146381, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 4.080884081824595e-05, | |
| "loss": 0.1514, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "learning_rate": 4.033858452856807e-05, | |
| "loss": 0.148, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 9.54, | |
| "learning_rate": 3.9868328238890195e-05, | |
| "loss": 0.1463, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 9.61, | |
| "learning_rate": 3.9398071949212326e-05, | |
| "loss": 0.1412, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 3.892781565953445e-05, | |
| "loss": 0.1479, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "eval_loss": 0.14136268198490143, | |
| "eval_runtime": 556.7019, | |
| "eval_samples_per_second": 9.766, | |
| "eval_steps_per_second": 0.611, | |
| "eval_wer": 0.11328894859148715, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "learning_rate": 3.8457559369856574e-05, | |
| "loss": 0.1541, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "learning_rate": 3.79873030801787e-05, | |
| "loss": 0.1482, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 3.751704679050083e-05, | |
| "loss": 0.1462, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "learning_rate": 3.705149306371973e-05, | |
| "loss": 0.1513, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "learning_rate": 3.6581236774041855e-05, | |
| "loss": 0.1607, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "eval_loss": 0.14584468305110931, | |
| "eval_runtime": 555.5836, | |
| "eval_samples_per_second": 9.786, | |
| "eval_steps_per_second": 0.612, | |
| "eval_wer": 0.11712277054918764, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 3.611098048436398e-05, | |
| "loss": 0.1589, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 10.12, | |
| "learning_rate": 3.56407241946861e-05, | |
| "loss": 0.1588, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 10.19, | |
| "learning_rate": 3.5170467905008234e-05, | |
| "loss": 0.1646, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 10.25, | |
| "learning_rate": 3.470021161533036e-05, | |
| "loss": 0.17, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 3.422995532565248e-05, | |
| "loss": 0.166, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "eval_loss": 0.16516457498073578, | |
| "eval_runtime": 557.4495, | |
| "eval_samples_per_second": 9.753, | |
| "eval_steps_per_second": 0.61, | |
| "eval_wer": 0.12636904703540647, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 3.3759699035974606e-05, | |
| "loss": 0.1784, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 10.44, | |
| "learning_rate": 3.328944274629674e-05, | |
| "loss": 0.1678, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 10.51, | |
| "learning_rate": 3.281918645661886e-05, | |
| "loss": 0.1771, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 10.57, | |
| "learning_rate": 3.235363272983776e-05, | |
| "loss": 0.1885, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 10.64, | |
| "learning_rate": 3.188337644015989e-05, | |
| "loss": 0.188, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 10.64, | |
| "eval_loss": 0.17125311493873596, | |
| "eval_runtime": 557.8968, | |
| "eval_samples_per_second": 9.746, | |
| "eval_steps_per_second": 0.609, | |
| "eval_wer": 0.13218351358506478, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 10.7, | |
| "learning_rate": 3.141312015048201e-05, | |
| "loss": 0.1847, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "learning_rate": 3.094286386080414e-05, | |
| "loss": 0.2217, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 10.83, | |
| "learning_rate": 3.0472607571126265e-05, | |
| "loss": 0.1662, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 10.9, | |
| "learning_rate": 3.0002351281448393e-05, | |
| "loss": 0.1461, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "learning_rate": 2.9532094991770514e-05, | |
| "loss": 0.1461, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "eval_loss": 0.14227142930030823, | |
| "eval_runtime": 556.6368, | |
| "eval_samples_per_second": 9.768, | |
| "eval_steps_per_second": 0.611, | |
| "eval_wer": 0.11110239540333572, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 2.906183870209264e-05, | |
| "loss": 0.1583, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 11.09, | |
| "learning_rate": 2.859158241241477e-05, | |
| "loss": 0.1221, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 11.15, | |
| "learning_rate": 2.812132612273689e-05, | |
| "loss": 0.138, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 11.22, | |
| "learning_rate": 2.7651069833059017e-05, | |
| "loss": 0.1269, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 11.28, | |
| "learning_rate": 2.7180813543381144e-05, | |
| "loss": 0.1289, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 11.28, | |
| "eval_loss": 0.13876527547836304, | |
| "eval_runtime": 555.2154, | |
| "eval_samples_per_second": 9.793, | |
| "eval_steps_per_second": 0.612, | |
| "eval_wer": 0.10974928177120613, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 11.35, | |
| "learning_rate": 2.671055725370327e-05, | |
| "loss": 0.1395, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 11.41, | |
| "learning_rate": 2.6240300964025392e-05, | |
| "loss": 0.1397, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 11.48, | |
| "learning_rate": 2.577004467434752e-05, | |
| "loss": 0.123, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 11.54, | |
| "learning_rate": 2.5299788384669647e-05, | |
| "loss": 0.1486, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "learning_rate": 2.482953209499177e-05, | |
| "loss": 0.1273, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "eval_loss": 0.1438097506761551, | |
| "eval_runtime": 554.8582, | |
| "eval_samples_per_second": 9.799, | |
| "eval_steps_per_second": 0.613, | |
| "eval_wer": 0.10742545618559228, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "learning_rate": 2.4363978368210676e-05, | |
| "loss": 0.1313, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 11.73, | |
| "learning_rate": 2.38937220785328e-05, | |
| "loss": 0.1365, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 11.8, | |
| "learning_rate": 2.3423465788854924e-05, | |
| "loss": 0.1335, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 2.295320949917705e-05, | |
| "loss": 0.1321, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "learning_rate": 2.2482953209499176e-05, | |
| "loss": 0.1317, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "eval_loss": 0.13120408356189728, | |
| "eval_runtime": 555.4622, | |
| "eval_samples_per_second": 9.788, | |
| "eval_steps_per_second": 0.612, | |
| "eval_wer": 0.10656260111582849, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "learning_rate": 2.2012696919821303e-05, | |
| "loss": 0.137, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 2.154244063014343e-05, | |
| "loss": 0.1224, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 12.12, | |
| "learning_rate": 2.1072184340465555e-05, | |
| "loss": 0.1293, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 12.19, | |
| "learning_rate": 2.0606630613684456e-05, | |
| "loss": 0.1245, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 12.25, | |
| "learning_rate": 2.0136374324006584e-05, | |
| "loss": 0.1448, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 12.25, | |
| "eval_loss": 0.1446371227502823, | |
| "eval_runtime": 558.424, | |
| "eval_samples_per_second": 9.736, | |
| "eval_steps_per_second": 0.609, | |
| "eval_wer": 0.10419955484522538, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 12.31, | |
| "learning_rate": 1.9666118034328708e-05, | |
| "loss": 0.1472, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 12.38, | |
| "learning_rate": 1.9195861744650835e-05, | |
| "loss": 0.15, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 12.44, | |
| "learning_rate": 1.8725605454972963e-05, | |
| "loss": 0.1434, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 12.51, | |
| "learning_rate": 1.8255349165295087e-05, | |
| "loss": 0.1439, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 12.57, | |
| "learning_rate": 1.7785092875617214e-05, | |
| "loss": 0.1424, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 12.57, | |
| "eval_loss": 0.13855019211769104, | |
| "eval_runtime": 555.0962, | |
| "eval_samples_per_second": 9.795, | |
| "eval_steps_per_second": 0.613, | |
| "eval_wer": 0.10152274309470814, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 12.64, | |
| "learning_rate": 1.7314836585939338e-05, | |
| "loss": 0.1422, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 12.7, | |
| "learning_rate": 1.6844580296261466e-05, | |
| "loss": 0.1426, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 12.77, | |
| "learning_rate": 1.637432400658359e-05, | |
| "loss": 0.1395, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 12.83, | |
| "learning_rate": 1.5904067716905717e-05, | |
| "loss": 0.1363, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 12.89, | |
| "learning_rate": 1.543381142722784e-05, | |
| "loss": 0.1392, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 12.89, | |
| "eval_loss": 0.13786287605762482, | |
| "eval_runtime": 553.5953, | |
| "eval_samples_per_second": 9.821, | |
| "eval_steps_per_second": 0.614, | |
| "eval_wer": 0.10046378459999804, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "learning_rate": 1.4963555137549964e-05, | |
| "loss": 0.1387, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 1.4493298847872091e-05, | |
| "loss": 0.1388, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 13.09, | |
| "learning_rate": 1.4023042558194215e-05, | |
| "loss": 0.1259, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 13.15, | |
| "learning_rate": 1.3552786268516343e-05, | |
| "loss": 0.1475, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 13.22, | |
| "learning_rate": 1.3082529978838467e-05, | |
| "loss": 0.1408, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 13.22, | |
| "eval_loss": 0.1407657265663147, | |
| "eval_runtime": 557.4847, | |
| "eval_samples_per_second": 9.753, | |
| "eval_steps_per_second": 0.61, | |
| "eval_wer": 0.09921852785158893, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 13.28, | |
| "learning_rate": 1.2612273689160592e-05, | |
| "loss": 0.1444, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 13.35, | |
| "learning_rate": 1.214201739948272e-05, | |
| "loss": 0.1324, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 13.41, | |
| "learning_rate": 1.1671761109804846e-05, | |
| "loss": 0.1397, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 13.48, | |
| "learning_rate": 1.120150482012697e-05, | |
| "loss": 0.1382, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 13.54, | |
| "learning_rate": 1.0731248530449096e-05, | |
| "loss": 0.1239, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 13.54, | |
| "eval_loss": 0.13379834592342377, | |
| "eval_runtime": 553.3736, | |
| "eval_samples_per_second": 9.825, | |
| "eval_steps_per_second": 0.614, | |
| "eval_wer": 0.09677704021100729, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "learning_rate": 1.0260992240771221e-05, | |
| "loss": 0.122, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 13.67, | |
| "learning_rate": 9.790735951093347e-06, | |
| "loss": 0.1209, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 13.73, | |
| "learning_rate": 9.320479661415473e-06, | |
| "loss": 0.1248, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 13.8, | |
| "learning_rate": 8.850223371737597e-06, | |
| "loss": 0.1198, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 13.86, | |
| "learning_rate": 8.379967082059723e-06, | |
| "loss": 0.1244, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 13.86, | |
| "eval_loss": 0.1335345059633255, | |
| "eval_runtime": 554.5112, | |
| "eval_samples_per_second": 9.805, | |
| "eval_steps_per_second": 0.613, | |
| "eval_wer": 0.09565925068881329, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "learning_rate": 7.914413355278628e-06, | |
| "loss": 0.1235, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "learning_rate": 7.444157065600752e-06, | |
| "loss": 0.1367, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 14.06, | |
| "learning_rate": 6.973900775922878e-06, | |
| "loss": 0.1288, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 14.12, | |
| "learning_rate": 6.503644486245005e-06, | |
| "loss": 0.1358, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 14.18, | |
| "learning_rate": 6.03338819656713e-06, | |
| "loss": 0.1254, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 14.18, | |
| "eval_loss": 0.13817058503627777, | |
| "eval_runtime": 555.1553, | |
| "eval_samples_per_second": 9.794, | |
| "eval_steps_per_second": 0.612, | |
| "eval_wer": 0.09503171972898507, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "learning_rate": 5.5631319068892555e-06, | |
| "loss": 0.1343, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 14.31, | |
| "learning_rate": 5.097578180108159e-06, | |
| "loss": 0.131, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 14.38, | |
| "learning_rate": 4.627321890430284e-06, | |
| "loss": 0.1503, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 14.44, | |
| "learning_rate": 4.15706560075241e-06, | |
| "loss": 0.1543, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 14.51, | |
| "learning_rate": 3.6868093110745355e-06, | |
| "loss": 0.1597, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 14.51, | |
| "eval_loss": 0.15444068610668182, | |
| "eval_runtime": 553.9922, | |
| "eval_samples_per_second": 9.814, | |
| "eval_steps_per_second": 0.614, | |
| "eval_wer": 0.09704177983468482, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 14.57, | |
| "learning_rate": 3.2165530213966613e-06, | |
| "loss": 0.1588, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 14.64, | |
| "learning_rate": 2.746296731718787e-06, | |
| "loss": 0.1616, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 14.7, | |
| "learning_rate": 2.2760404420409124e-06, | |
| "loss": 0.1701, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 14.76, | |
| "learning_rate": 1.8057841523630381e-06, | |
| "loss": 0.1569, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 14.83, | |
| "learning_rate": 1.3355278626851635e-06, | |
| "loss": 0.1566, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 14.83, | |
| "eval_loss": 0.1588866114616394, | |
| "eval_runtime": 554.1492, | |
| "eval_samples_per_second": 9.811, | |
| "eval_steps_per_second": 0.614, | |
| "eval_wer": 0.09631619716238345, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 14.89, | |
| "learning_rate": 8.652715730072889e-07, | |
| "loss": 0.1645, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 14.96, | |
| "learning_rate": 3.950152833294146e-07, | |
| "loss": 0.1684, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 23265, | |
| "total_flos": 4.936819874574462e+20, | |
| "train_loss": 0.2744614067626927, | |
| "train_runtime": 128452.941, | |
| "train_samples_per_second": 5.797, | |
| "train_steps_per_second": 0.181 | |
| } | |
| ], | |
| "max_steps": 23265, | |
| "num_train_epochs": 15, | |
| "total_flos": 4.936819874574462e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |