| { | |
| "best_metric": 6.544273760459599, | |
| "best_model_checkpoint": "./checkpoint-19500", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 20000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00125, | |
| "grad_norm": 10.480854034423828, | |
| "learning_rate": 1.0499999999999999e-07, | |
| "loss": 1.4117, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0025, | |
| "grad_norm": 9.908479690551758, | |
| "learning_rate": 2.1437499999999999e-07, | |
| "loss": 1.4142, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.00375, | |
| "grad_norm": 7.737735748291016, | |
| "learning_rate": 3.2374999999999997e-07, | |
| "loss": 1.2755, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.005, | |
| "grad_norm": 5.847314834594727, | |
| "learning_rate": 4.33125e-07, | |
| "loss": 0.8497, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.00625, | |
| "grad_norm": 5.064878463745117, | |
| "learning_rate": 5.425e-07, | |
| "loss": 0.5289, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.0075, | |
| "grad_norm": 4.789758205413818, | |
| "learning_rate": 6.518749999999999e-07, | |
| "loss": 0.5181, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.00875, | |
| "grad_norm": 4.744896411895752, | |
| "learning_rate": 7.612499999999999e-07, | |
| "loss": 0.3765, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 4.427361965179443, | |
| "learning_rate": 8.706249999999999e-07, | |
| "loss": 0.3585, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.01125, | |
| "grad_norm": 4.670985698699951, | |
| "learning_rate": 9.8e-07, | |
| "loss": 0.3219, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.0125, | |
| "grad_norm": 3.8691747188568115, | |
| "learning_rate": 1.0893749999999998e-06, | |
| "loss": 0.3409, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.01375, | |
| "grad_norm": 4.622318267822266, | |
| "learning_rate": 1.19875e-06, | |
| "loss": 0.2899, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.015, | |
| "grad_norm": 3.7900593280792236, | |
| "learning_rate": 1.3081249999999999e-06, | |
| "loss": 0.2834, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.01625, | |
| "grad_norm": 3.9323770999908447, | |
| "learning_rate": 1.4175e-06, | |
| "loss": 0.2643, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.0175, | |
| "grad_norm": 3.708969831466675, | |
| "learning_rate": 1.5268749999999997e-06, | |
| "loss": 0.2827, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.01875, | |
| "grad_norm": 3.707580804824829, | |
| "learning_rate": 1.6362499999999998e-06, | |
| "loss": 0.274, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 3.8519480228424072, | |
| "learning_rate": 1.745625e-06, | |
| "loss": 0.2568, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.02125, | |
| "grad_norm": 4.571149826049805, | |
| "learning_rate": 1.8549999999999998e-06, | |
| "loss": 0.2376, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.0225, | |
| "grad_norm": 3.2112503051757812, | |
| "learning_rate": 1.9643749999999997e-06, | |
| "loss": 0.2212, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.02375, | |
| "grad_norm": 6.173221588134766, | |
| "learning_rate": 2.07375e-06, | |
| "loss": 0.2238, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.025, | |
| "grad_norm": 4.781201362609863, | |
| "learning_rate": 2.183125e-06, | |
| "loss": 0.2854, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.025, | |
| "eval_loss": 0.41938766837120056, | |
| "eval_runtime": 531.5621, | |
| "eval_samples_per_second": 3.181, | |
| "eval_steps_per_second": 0.399, | |
| "eval_wer": 25.88984638441364, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.02625, | |
| "grad_norm": 4.559605598449707, | |
| "learning_rate": 2.2925e-06, | |
| "loss": 0.2996, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.0275, | |
| "grad_norm": 5.1446852684021, | |
| "learning_rate": 2.401875e-06, | |
| "loss": 0.2791, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.02875, | |
| "grad_norm": 4.178796768188477, | |
| "learning_rate": 2.5112499999999995e-06, | |
| "loss": 0.3061, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 4.030816555023193, | |
| "learning_rate": 2.6206249999999996e-06, | |
| "loss": 0.2768, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.03125, | |
| "grad_norm": 4.405904769897461, | |
| "learning_rate": 2.7299999999999997e-06, | |
| "loss": 0.2678, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.0325, | |
| "grad_norm": 3.5179622173309326, | |
| "learning_rate": 2.839375e-06, | |
| "loss": 0.2363, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.03375, | |
| "grad_norm": 3.767529010772705, | |
| "learning_rate": 2.94875e-06, | |
| "loss": 0.219, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.035, | |
| "grad_norm": 2.898439645767212, | |
| "learning_rate": 3.0581249999999996e-06, | |
| "loss": 0.1913, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.03625, | |
| "grad_norm": 3.3581595420837402, | |
| "learning_rate": 3.1674999999999997e-06, | |
| "loss": 0.173, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.0375, | |
| "grad_norm": 2.872340202331543, | |
| "learning_rate": 3.276875e-06, | |
| "loss": 0.176, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.03875, | |
| "grad_norm": 3.1199734210968018, | |
| "learning_rate": 3.38625e-06, | |
| "loss": 0.1808, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 2.626908779144287, | |
| "learning_rate": 3.495625e-06, | |
| "loss": 0.1802, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.04125, | |
| "grad_norm": 2.776141881942749, | |
| "learning_rate": 3.6049999999999994e-06, | |
| "loss": 0.1761, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.0425, | |
| "grad_norm": 3.149322509765625, | |
| "learning_rate": 3.7143749999999995e-06, | |
| "loss": 0.1569, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.04375, | |
| "grad_norm": 2.776301145553589, | |
| "learning_rate": 3.82375e-06, | |
| "loss": 0.1607, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.045, | |
| "grad_norm": 2.8884706497192383, | |
| "learning_rate": 3.933125e-06, | |
| "loss": 0.1582, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.04625, | |
| "grad_norm": 3.991647720336914, | |
| "learning_rate": 4.0425e-06, | |
| "loss": 0.1511, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.0475, | |
| "grad_norm": 2.892364740371704, | |
| "learning_rate": 4.151874999999999e-06, | |
| "loss": 0.1434, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.04875, | |
| "grad_norm": 3.7825698852539062, | |
| "learning_rate": 4.261249999999999e-06, | |
| "loss": 0.148, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 2.511293649673462, | |
| "learning_rate": 4.370624999999999e-06, | |
| "loss": 0.1425, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 0.39225178956985474, | |
| "eval_runtime": 528.4404, | |
| "eval_samples_per_second": 3.2, | |
| "eval_steps_per_second": 0.401, | |
| "eval_wer": 20.507056325715002, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.05125, | |
| "grad_norm": 1.8921111822128296, | |
| "learning_rate": 4.369473684210526e-06, | |
| "loss": 0.1395, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.0525, | |
| "grad_norm": 3.18829607963562, | |
| "learning_rate": 4.363717105263158e-06, | |
| "loss": 0.1294, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.05375, | |
| "grad_norm": 2.508878231048584, | |
| "learning_rate": 4.357960526315789e-06, | |
| "loss": 0.1366, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.055, | |
| "grad_norm": 2.201958179473877, | |
| "learning_rate": 4.352203947368421e-06, | |
| "loss": 0.14, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.05625, | |
| "grad_norm": 2.756673574447632, | |
| "learning_rate": 4.346447368421052e-06, | |
| "loss": 0.1355, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.0575, | |
| "grad_norm": 3.084169864654541, | |
| "learning_rate": 4.340690789473684e-06, | |
| "loss": 0.1278, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.05875, | |
| "grad_norm": 2.486377239227295, | |
| "learning_rate": 4.334934210526315e-06, | |
| "loss": 0.1298, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 3.5559706687927246, | |
| "learning_rate": 4.329177631578947e-06, | |
| "loss": 0.1352, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.06125, | |
| "grad_norm": 2.6353018283843994, | |
| "learning_rate": 4.323421052631579e-06, | |
| "loss": 0.1163, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 0.0625, | |
| "grad_norm": 2.8629567623138428, | |
| "learning_rate": 4.31766447368421e-06, | |
| "loss": 0.1199, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.06375, | |
| "grad_norm": 2.9020206928253174, | |
| "learning_rate": 4.311907894736842e-06, | |
| "loss": 0.1206, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 0.065, | |
| "grad_norm": 2.4626991748809814, | |
| "learning_rate": 4.306151315789473e-06, | |
| "loss": 0.1395, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.06625, | |
| "grad_norm": 2.9234840869903564, | |
| "learning_rate": 4.300394736842105e-06, | |
| "loss": 0.1269, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 0.0675, | |
| "grad_norm": 3.017625570297241, | |
| "learning_rate": 4.294638157894737e-06, | |
| "loss": 0.1228, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.06875, | |
| "grad_norm": 2.5392937660217285, | |
| "learning_rate": 4.288881578947368e-06, | |
| "loss": 0.1273, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 3.2523694038391113, | |
| "learning_rate": 4.283125e-06, | |
| "loss": 0.1242, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.07125, | |
| "grad_norm": 2.633652448654175, | |
| "learning_rate": 4.277368421052632e-06, | |
| "loss": 0.1341, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 0.0725, | |
| "grad_norm": 3.952681064605713, | |
| "learning_rate": 4.271611842105263e-06, | |
| "loss": 0.1588, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.07375, | |
| "grad_norm": 3.9815685749053955, | |
| "learning_rate": 4.265855263157895e-06, | |
| "loss": 0.1879, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 0.075, | |
| "grad_norm": 3.197030544281006, | |
| "learning_rate": 4.260098684210526e-06, | |
| "loss": 0.2199, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.075, | |
| "eval_loss": 0.3290639817714691, | |
| "eval_runtime": 535.231, | |
| "eval_samples_per_second": 3.159, | |
| "eval_steps_per_second": 0.396, | |
| "eval_wer": 17.478456350693143, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.07625, | |
| "grad_norm": 3.8294057846069336, | |
| "learning_rate": 4.254342105263158e-06, | |
| "loss": 0.205, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 0.0775, | |
| "grad_norm": 3.339564085006714, | |
| "learning_rate": 4.248585526315789e-06, | |
| "loss": 0.1793, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.07875, | |
| "grad_norm": 4.41719913482666, | |
| "learning_rate": 4.242828947368421e-06, | |
| "loss": 0.1903, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 4.329945087432861, | |
| "learning_rate": 4.237072368421052e-06, | |
| "loss": 0.2487, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.08125, | |
| "grad_norm": 2.858635425567627, | |
| "learning_rate": 4.231315789473684e-06, | |
| "loss": 0.16, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 0.0825, | |
| "grad_norm": 2.6474554538726807, | |
| "learning_rate": 4.225559210526316e-06, | |
| "loss": 0.1294, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.08375, | |
| "grad_norm": 2.6311450004577637, | |
| "learning_rate": 4.219802631578947e-06, | |
| "loss": 0.1199, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 0.085, | |
| "grad_norm": 2.472925901412964, | |
| "learning_rate": 4.214046052631579e-06, | |
| "loss": 0.1106, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.08625, | |
| "grad_norm": 2.1684815883636475, | |
| "learning_rate": 4.20828947368421e-06, | |
| "loss": 0.1081, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 0.0875, | |
| "grad_norm": 2.2405142784118652, | |
| "learning_rate": 4.202532894736842e-06, | |
| "loss": 0.1024, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.08875, | |
| "grad_norm": 3.28480863571167, | |
| "learning_rate": 4.196776315789474e-06, | |
| "loss": 0.1395, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 2.734311819076538, | |
| "learning_rate": 4.191019736842105e-06, | |
| "loss": 0.1685, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.09125, | |
| "grad_norm": 3.3384852409362793, | |
| "learning_rate": 4.185263157894737e-06, | |
| "loss": 0.179, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 0.0925, | |
| "grad_norm": 4.151054859161377, | |
| "learning_rate": 4.1795065789473686e-06, | |
| "loss": 0.1863, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.09375, | |
| "grad_norm": 3.854214668273926, | |
| "learning_rate": 4.17375e-06, | |
| "loss": 0.1753, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 0.095, | |
| "grad_norm": 3.3321709632873535, | |
| "learning_rate": 4.1679934210526316e-06, | |
| "loss": 0.1684, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.09625, | |
| "grad_norm": 2.8302998542785645, | |
| "learning_rate": 4.162236842105263e-06, | |
| "loss": 0.1818, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 0.0975, | |
| "grad_norm": 5.4296555519104, | |
| "learning_rate": 4.156480263157895e-06, | |
| "loss": 0.5523, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.09875, | |
| "grad_norm": 3.8675997257232666, | |
| "learning_rate": 4.1507236842105265e-06, | |
| "loss": 0.3352, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 3.9055581092834473, | |
| "learning_rate": 4.144967105263158e-06, | |
| "loss": 0.2343, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.2860749065876007, | |
| "eval_runtime": 530.1633, | |
| "eval_samples_per_second": 3.19, | |
| "eval_steps_per_second": 0.4, | |
| "eval_wer": 14.13138503809167, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.10125, | |
| "grad_norm": 3.413243293762207, | |
| "learning_rate": 4.1392105263157895e-06, | |
| "loss": 0.2233, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 0.1025, | |
| "grad_norm": 3.1129419803619385, | |
| "learning_rate": 4.133453947368421e-06, | |
| "loss": 0.2076, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.10375, | |
| "grad_norm": 3.0855767726898193, | |
| "learning_rate": 4.1276973684210525e-06, | |
| "loss": 0.1675, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 0.105, | |
| "grad_norm": 2.5053539276123047, | |
| "learning_rate": 4.121940789473684e-06, | |
| "loss": 0.1291, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.10625, | |
| "grad_norm": 2.078958511352539, | |
| "learning_rate": 4.1161842105263155e-06, | |
| "loss": 0.1036, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 0.1075, | |
| "grad_norm": 2.436898708343506, | |
| "learning_rate": 4.110427631578947e-06, | |
| "loss": 0.1153, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.10875, | |
| "grad_norm": 2.3834900856018066, | |
| "learning_rate": 4.1046710526315786e-06, | |
| "loss": 0.0943, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 2.070406913757324, | |
| "learning_rate": 4.09891447368421e-06, | |
| "loss": 0.0898, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.11125, | |
| "grad_norm": 2.0004026889801025, | |
| "learning_rate": 4.0931578947368416e-06, | |
| "loss": 0.0912, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 0.1125, | |
| "grad_norm": 2.4464359283447266, | |
| "learning_rate": 4.0874013157894735e-06, | |
| "loss": 0.0907, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.11375, | |
| "grad_norm": 2.8847742080688477, | |
| "learning_rate": 4.081644736842105e-06, | |
| "loss": 0.0978, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 0.115, | |
| "grad_norm": 2.167893171310425, | |
| "learning_rate": 4.0758881578947365e-06, | |
| "loss": 0.0968, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.11625, | |
| "grad_norm": 1.575804352760315, | |
| "learning_rate": 4.0701315789473684e-06, | |
| "loss": 0.0976, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 0.1175, | |
| "grad_norm": 2.3064370155334473, | |
| "learning_rate": 4.0643749999999995e-06, | |
| "loss": 0.0966, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.11875, | |
| "grad_norm": 1.9859708547592163, | |
| "learning_rate": 4.0586184210526314e-06, | |
| "loss": 0.0958, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 1.7534265518188477, | |
| "learning_rate": 4.052861842105263e-06, | |
| "loss": 0.0966, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.12125, | |
| "grad_norm": 3.8807549476623535, | |
| "learning_rate": 4.0471052631578945e-06, | |
| "loss": 0.1367, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 0.1225, | |
| "grad_norm": 3.933382987976074, | |
| "learning_rate": 4.041348684210526e-06, | |
| "loss": 0.1445, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.12375, | |
| "grad_norm": 3.3107643127441406, | |
| "learning_rate": 4.0355921052631575e-06, | |
| "loss": 0.1486, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 2.699190139770508, | |
| "learning_rate": 4.029835526315789e-06, | |
| "loss": 0.1391, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "eval_loss": 0.2906411290168762, | |
| "eval_runtime": 531.8376, | |
| "eval_samples_per_second": 3.18, | |
| "eval_steps_per_second": 0.399, | |
| "eval_wer": 13.31335081803422, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.12625, | |
| "grad_norm": 3.5841128826141357, | |
| "learning_rate": 4.0240789473684205e-06, | |
| "loss": 0.1628, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 0.1275, | |
| "grad_norm": 3.2463104724884033, | |
| "learning_rate": 4.018322368421052e-06, | |
| "loss": 0.1705, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.12875, | |
| "grad_norm": 5.094871520996094, | |
| "learning_rate": 4.0125657894736835e-06, | |
| "loss": 0.1663, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 3.0208804607391357, | |
| "learning_rate": 4.0068092105263154e-06, | |
| "loss": 0.1712, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.13125, | |
| "grad_norm": 3.4919967651367188, | |
| "learning_rate": 4.0010526315789465e-06, | |
| "loss": 0.1498, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 0.1325, | |
| "grad_norm": 3.4352219104766846, | |
| "learning_rate": 3.9952960526315784e-06, | |
| "loss": 0.1423, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.13375, | |
| "grad_norm": 3.63608455657959, | |
| "learning_rate": 3.98953947368421e-06, | |
| "loss": 0.1565, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 0.135, | |
| "grad_norm": 3.6155622005462646, | |
| "learning_rate": 3.9837828947368414e-06, | |
| "loss": 0.1553, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.13625, | |
| "grad_norm": 3.4833076000213623, | |
| "learning_rate": 3.978026315789473e-06, | |
| "loss": 0.1416, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 0.1375, | |
| "grad_norm": 3.147080421447754, | |
| "learning_rate": 3.9722697368421045e-06, | |
| "loss": 0.1469, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.13875, | |
| "grad_norm": 2.891146659851074, | |
| "learning_rate": 3.966513157894736e-06, | |
| "loss": 0.128, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 3.1411890983581543, | |
| "learning_rate": 3.960756578947368e-06, | |
| "loss": 0.1372, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.14125, | |
| "grad_norm": 3.836360216140747, | |
| "learning_rate": 3.954999999999999e-06, | |
| "loss": 0.149, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 0.1425, | |
| "grad_norm": 4.1377339363098145, | |
| "learning_rate": 3.949243421052631e-06, | |
| "loss": 0.1383, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.14375, | |
| "grad_norm": 3.7640268802642822, | |
| "learning_rate": 3.943486842105263e-06, | |
| "loss": 0.1492, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 0.145, | |
| "grad_norm": 3.452561855316162, | |
| "learning_rate": 3.937730263157894e-06, | |
| "loss": 0.1288, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.14625, | |
| "grad_norm": 1.9712022542953491, | |
| "learning_rate": 3.931973684210526e-06, | |
| "loss": 0.1137, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 0.1475, | |
| "grad_norm": 2.0524768829345703, | |
| "learning_rate": 3.926217105263157e-06, | |
| "loss": 0.0973, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.14875, | |
| "grad_norm": 2.3722898960113525, | |
| "learning_rate": 3.920460526315789e-06, | |
| "loss": 0.0915, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 2.5048534870147705, | |
| "learning_rate": 3.914703947368421e-06, | |
| "loss": 0.0853, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 0.26879894733428955, | |
| "eval_runtime": 531.192, | |
| "eval_samples_per_second": 3.183, | |
| "eval_steps_per_second": 0.399, | |
| "eval_wer": 12.045710003746722, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.15125, | |
| "grad_norm": 2.009464740753174, | |
| "learning_rate": 3.908947368421052e-06, | |
| "loss": 0.1037, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 0.1525, | |
| "grad_norm": 2.3635034561157227, | |
| "learning_rate": 3.903190789473684e-06, | |
| "loss": 0.0889, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.15375, | |
| "grad_norm": 3.131683111190796, | |
| "learning_rate": 3.897434210526315e-06, | |
| "loss": 0.0895, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 0.155, | |
| "grad_norm": 2.2032673358917236, | |
| "learning_rate": 3.891677631578947e-06, | |
| "loss": 0.0955, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.15625, | |
| "grad_norm": 1.8079180717468262, | |
| "learning_rate": 3.885921052631578e-06, | |
| "loss": 0.0785, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 0.1575, | |
| "grad_norm": 2.2879910469055176, | |
| "learning_rate": 3.88016447368421e-06, | |
| "loss": 0.0721, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.15875, | |
| "grad_norm": 2.491487979888916, | |
| "learning_rate": 3.874407894736841e-06, | |
| "loss": 0.0817, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 2.996129035949707, | |
| "learning_rate": 3.868651315789473e-06, | |
| "loss": 0.0877, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.16125, | |
| "grad_norm": 2.9992258548736572, | |
| "learning_rate": 3.862894736842104e-06, | |
| "loss": 0.1011, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 0.1625, | |
| "grad_norm": 2.5464529991149902, | |
| "learning_rate": 3.857138157894736e-06, | |
| "loss": 0.1385, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.16375, | |
| "grad_norm": 2.853933095932007, | |
| "learning_rate": 3.851381578947368e-06, | |
| "loss": 0.1223, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 0.165, | |
| "grad_norm": 2.3290011882781982, | |
| "learning_rate": 3.845624999999999e-06, | |
| "loss": 0.1236, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.16625, | |
| "grad_norm": 2.61714768409729, | |
| "learning_rate": 3.839868421052631e-06, | |
| "loss": 0.1167, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 0.1675, | |
| "grad_norm": 3.926612377166748, | |
| "learning_rate": 3.834111842105263e-06, | |
| "loss": 0.1306, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 0.16875, | |
| "grad_norm": 2.9979617595672607, | |
| "learning_rate": 3.828355263157894e-06, | |
| "loss": 0.1383, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 2.879436492919922, | |
| "learning_rate": 3.822598684210526e-06, | |
| "loss": 0.1112, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.17125, | |
| "grad_norm": 2.706355571746826, | |
| "learning_rate": 3.816842105263158e-06, | |
| "loss": 0.1006, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 0.1725, | |
| "grad_norm": 2.263953685760498, | |
| "learning_rate": 3.811085526315789e-06, | |
| "loss": 0.089, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 0.17375, | |
| "grad_norm": 3.070748805999756, | |
| "learning_rate": 3.8053289473684207e-06, | |
| "loss": 0.0801, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 0.175, | |
| "grad_norm": 2.202629566192627, | |
| "learning_rate": 3.799572368421052e-06, | |
| "loss": 0.0866, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.175, | |
| "eval_loss": 0.2575243413448334, | |
| "eval_runtime": 531.9323, | |
| "eval_samples_per_second": 3.179, | |
| "eval_steps_per_second": 0.399, | |
| "eval_wer": 11.471212688897214, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.17625, | |
| "grad_norm": 2.1003735065460205, | |
| "learning_rate": 3.793815789473684e-06, | |
| "loss": 0.0794, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 0.1775, | |
| "grad_norm": 2.4936602115631104, | |
| "learning_rate": 3.788059210526315e-06, | |
| "loss": 0.0754, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 0.17875, | |
| "grad_norm": 2.2320945262908936, | |
| "learning_rate": 3.782302631578947e-06, | |
| "loss": 0.0906, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 1.4985826015472412, | |
| "learning_rate": 3.7765460526315786e-06, | |
| "loss": 0.0872, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.18125, | |
| "grad_norm": 1.6096969842910767, | |
| "learning_rate": 3.77078947368421e-06, | |
| "loss": 0.0783, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 0.1825, | |
| "grad_norm": 1.8306738138198853, | |
| "learning_rate": 3.7650328947368416e-06, | |
| "loss": 0.0783, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 0.18375, | |
| "grad_norm": 1.972235918045044, | |
| "learning_rate": 3.759276315789473e-06, | |
| "loss": 0.0789, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 0.185, | |
| "grad_norm": 2.0266051292419434, | |
| "learning_rate": 3.753519736842105e-06, | |
| "loss": 0.0708, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.18625, | |
| "grad_norm": 1.6395690441131592, | |
| "learning_rate": 3.7477631578947366e-06, | |
| "loss": 0.0768, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 0.1875, | |
| "grad_norm": 1.8886572122573853, | |
| "learning_rate": 3.742006578947368e-06, | |
| "loss": 0.0745, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.18875, | |
| "grad_norm": 3.0105178356170654, | |
| "learning_rate": 3.7362499999999996e-06, | |
| "loss": 0.0816, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 1.8376508951187134, | |
| "learning_rate": 3.7304934210526315e-06, | |
| "loss": 0.082, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.19125, | |
| "grad_norm": 1.758370041847229, | |
| "learning_rate": 3.7247368421052626e-06, | |
| "loss": 0.0798, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 0.1925, | |
| "grad_norm": 1.2405736446380615, | |
| "learning_rate": 3.7189802631578945e-06, | |
| "loss": 0.0773, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 0.19375, | |
| "grad_norm": 1.8085663318634033, | |
| "learning_rate": 3.713223684210526e-06, | |
| "loss": 0.0861, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 0.195, | |
| "grad_norm": 3.838613986968994, | |
| "learning_rate": 3.7074671052631575e-06, | |
| "loss": 0.1032, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.19625, | |
| "grad_norm": 3.087472438812256, | |
| "learning_rate": 3.701710526315789e-06, | |
| "loss": 0.133, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 0.1975, | |
| "grad_norm": 2.3854024410247803, | |
| "learning_rate": 3.6959539473684206e-06, | |
| "loss": 0.1366, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 0.19875, | |
| "grad_norm": 3.235400676727295, | |
| "learning_rate": 3.690197368421052e-06, | |
| "loss": 0.1275, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 2.7316720485687256, | |
| "learning_rate": 3.684440789473684e-06, | |
| "loss": 0.1311, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.24720044434070587, | |
| "eval_runtime": 536.8744, | |
| "eval_samples_per_second": 3.15, | |
| "eval_steps_per_second": 0.395, | |
| "eval_wer": 12.482827525914825, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.20125, | |
| "grad_norm": 2.4120874404907227, | |
| "learning_rate": 3.6786842105263155e-06, | |
| "loss": 0.1215, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 0.2025, | |
| "grad_norm": 2.5485270023345947, | |
| "learning_rate": 3.672927631578947e-06, | |
| "loss": 0.0983, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 0.20375, | |
| "grad_norm": 2.2741594314575195, | |
| "learning_rate": 3.667171052631579e-06, | |
| "loss": 0.0764, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 0.205, | |
| "grad_norm": 1.875857949256897, | |
| "learning_rate": 3.66141447368421e-06, | |
| "loss": 0.0733, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.20625, | |
| "grad_norm": 1.8897082805633545, | |
| "learning_rate": 3.655657894736842e-06, | |
| "loss": 0.0797, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 0.2075, | |
| "grad_norm": 1.5462270975112915, | |
| "learning_rate": 3.6499013157894735e-06, | |
| "loss": 0.0772, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 0.20875, | |
| "grad_norm": 2.1055002212524414, | |
| "learning_rate": 3.644144736842105e-06, | |
| "loss": 0.079, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 2.8036248683929443, | |
| "learning_rate": 3.6383881578947365e-06, | |
| "loss": 0.0828, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.21125, | |
| "grad_norm": 1.496777892112732, | |
| "learning_rate": 3.6326315789473684e-06, | |
| "loss": 0.0658, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 0.2125, | |
| "grad_norm": 2.213822364807129, | |
| "learning_rate": 3.6268749999999995e-06, | |
| "loss": 0.0722, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.21375, | |
| "grad_norm": 1.4431771039962769, | |
| "learning_rate": 3.6211184210526314e-06, | |
| "loss": 0.061, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 0.215, | |
| "grad_norm": 1.6346482038497925, | |
| "learning_rate": 3.6153618421052625e-06, | |
| "loss": 0.0641, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.21625, | |
| "grad_norm": 1.5905380249023438, | |
| "learning_rate": 3.6096052631578944e-06, | |
| "loss": 0.0633, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 0.2175, | |
| "grad_norm": 2.4848458766937256, | |
| "learning_rate": 3.6038486842105263e-06, | |
| "loss": 0.0738, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 0.21875, | |
| "grad_norm": 2.568466901779175, | |
| "learning_rate": 3.5980921052631574e-06, | |
| "loss": 0.1123, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 2.5104339122772217, | |
| "learning_rate": 3.5923355263157894e-06, | |
| "loss": 0.1179, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.22125, | |
| "grad_norm": 3.769829273223877, | |
| "learning_rate": 3.586578947368421e-06, | |
| "loss": 0.1221, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 0.2225, | |
| "grad_norm": 2.850048542022705, | |
| "learning_rate": 3.5808223684210524e-06, | |
| "loss": 0.1115, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 0.22375, | |
| "grad_norm": 2.0328500270843506, | |
| "learning_rate": 3.575065789473684e-06, | |
| "loss": 0.1274, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 0.225, | |
| "grad_norm": 2.765300750732422, | |
| "learning_rate": 3.569309210526316e-06, | |
| "loss": 0.1338, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.225, | |
| "eval_loss": 0.24367325007915497, | |
| "eval_runtime": 531.4688, | |
| "eval_samples_per_second": 3.182, | |
| "eval_steps_per_second": 0.399, | |
| "eval_wer": 10.990383414512301, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.22625, | |
| "grad_norm": 1.4456897974014282, | |
| "learning_rate": 3.563552631578947e-06, | |
| "loss": 0.0921, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 0.2275, | |
| "grad_norm": 2.357384443283081, | |
| "learning_rate": 3.557796052631579e-06, | |
| "loss": 0.0728, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 0.22875, | |
| "grad_norm": 2.2841663360595703, | |
| "learning_rate": 3.55203947368421e-06, | |
| "loss": 0.0703, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 1.8975858688354492, | |
| "learning_rate": 3.546282894736842e-06, | |
| "loss": 0.0595, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.23125, | |
| "grad_norm": 1.6614043712615967, | |
| "learning_rate": 3.5405263157894733e-06, | |
| "loss": 0.0684, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 0.2325, | |
| "grad_norm": 3.0987887382507324, | |
| "learning_rate": 3.534769736842105e-06, | |
| "loss": 0.0643, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 0.23375, | |
| "grad_norm": 1.869446873664856, | |
| "learning_rate": 3.5290131578947363e-06, | |
| "loss": 0.0612, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 0.235, | |
| "grad_norm": 1.6360236406326294, | |
| "learning_rate": 3.5232565789473683e-06, | |
| "loss": 0.0627, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.23625, | |
| "grad_norm": 2.188901424407959, | |
| "learning_rate": 3.5174999999999998e-06, | |
| "loss": 0.068, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 0.2375, | |
| "grad_norm": 1.5851141214370728, | |
| "learning_rate": 3.5117434210526313e-06, | |
| "loss": 0.0702, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 0.23875, | |
| "grad_norm": 1.9303579330444336, | |
| "learning_rate": 3.5059868421052632e-06, | |
| "loss": 0.0683, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 1.8640798330307007, | |
| "learning_rate": 3.5002302631578943e-06, | |
| "loss": 0.0637, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.24125, | |
| "grad_norm": 2.395669937133789, | |
| "learning_rate": 3.4944736842105262e-06, | |
| "loss": 0.0626, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 0.2425, | |
| "grad_norm": 1.5368024110794067, | |
| "learning_rate": 3.4887171052631573e-06, | |
| "loss": 0.0694, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 0.24375, | |
| "grad_norm": 2.1346402168273926, | |
| "learning_rate": 3.4829605263157892e-06, | |
| "loss": 0.0734, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 0.245, | |
| "grad_norm": 2.0883893966674805, | |
| "learning_rate": 3.4772039473684207e-06, | |
| "loss": 0.0659, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.24625, | |
| "grad_norm": 1.6861238479614258, | |
| "learning_rate": 3.4714473684210523e-06, | |
| "loss": 0.0656, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 0.2475, | |
| "grad_norm": 1.5790470838546753, | |
| "learning_rate": 3.4656907894736838e-06, | |
| "loss": 0.0801, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 0.24875, | |
| "grad_norm": 1.3223644495010376, | |
| "learning_rate": 3.4599342105263157e-06, | |
| "loss": 0.0806, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 1.6931387186050415, | |
| "learning_rate": 3.4541776315789468e-06, | |
| "loss": 0.0748, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.2556721270084381, | |
| "eval_runtime": 534.5469, | |
| "eval_samples_per_second": 3.163, | |
| "eval_steps_per_second": 0.397, | |
| "eval_wer": 10.709379293118522, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.25125, | |
| "grad_norm": 1.5327143669128418, | |
| "learning_rate": 3.4484210526315787e-06, | |
| "loss": 0.0968, | |
| "step": 5025 | |
| }, | |
| { | |
| "epoch": 0.2525, | |
| "grad_norm": 1.988226294517517, | |
| "learning_rate": 3.4426644736842106e-06, | |
| "loss": 0.0921, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 0.25375, | |
| "grad_norm": 2.179086446762085, | |
| "learning_rate": 3.4369078947368417e-06, | |
| "loss": 0.093, | |
| "step": 5075 | |
| }, | |
| { | |
| "epoch": 0.255, | |
| "grad_norm": 2.4304797649383545, | |
| "learning_rate": 3.4311513157894736e-06, | |
| "loss": 0.0909, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 0.25625, | |
| "grad_norm": 2.498908281326294, | |
| "learning_rate": 3.4253947368421047e-06, | |
| "loss": 0.1225, | |
| "step": 5125 | |
| }, | |
| { | |
| "epoch": 0.2575, | |
| "grad_norm": 2.018110752105713, | |
| "learning_rate": 3.4196381578947367e-06, | |
| "loss": 0.1199, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 0.25875, | |
| "grad_norm": 1.8156744241714478, | |
| "learning_rate": 3.413881578947368e-06, | |
| "loss": 0.1032, | |
| "step": 5175 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 2.395634651184082, | |
| "learning_rate": 3.4081249999999997e-06, | |
| "loss": 0.0842, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.26125, | |
| "grad_norm": 1.8604170083999634, | |
| "learning_rate": 3.402368421052631e-06, | |
| "loss": 0.0753, | |
| "step": 5225 | |
| }, | |
| { | |
| "epoch": 0.2625, | |
| "grad_norm": 2.186006784439087, | |
| "learning_rate": 3.396611842105263e-06, | |
| "loss": 0.0693, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 0.26375, | |
| "grad_norm": 2.117950201034546, | |
| "learning_rate": 3.390855263157894e-06, | |
| "loss": 0.0731, | |
| "step": 5275 | |
| }, | |
| { | |
| "epoch": 0.265, | |
| "grad_norm": 1.442688226699829, | |
| "learning_rate": 3.385098684210526e-06, | |
| "loss": 0.0607, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 0.26625, | |
| "grad_norm": 2.0623013973236084, | |
| "learning_rate": 3.379342105263157e-06, | |
| "loss": 0.0598, | |
| "step": 5325 | |
| }, | |
| { | |
| "epoch": 0.2675, | |
| "grad_norm": 1.6096211671829224, | |
| "learning_rate": 3.373585526315789e-06, | |
| "loss": 0.0687, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 0.26875, | |
| "grad_norm": 1.2381603717803955, | |
| "learning_rate": 3.367828947368421e-06, | |
| "loss": 0.0646, | |
| "step": 5375 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 1.6694140434265137, | |
| "learning_rate": 3.362072368421052e-06, | |
| "loss": 0.0595, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.27125, | |
| "grad_norm": 2.486950159072876, | |
| "learning_rate": 3.356315789473684e-06, | |
| "loss": 0.074, | |
| "step": 5425 | |
| }, | |
| { | |
| "epoch": 0.2725, | |
| "grad_norm": 1.2931033372879028, | |
| "learning_rate": 3.3505592105263156e-06, | |
| "loss": 0.08, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 0.27375, | |
| "grad_norm": 2.314680337905884, | |
| "learning_rate": 3.344802631578947e-06, | |
| "loss": 0.0662, | |
| "step": 5475 | |
| }, | |
| { | |
| "epoch": 0.275, | |
| "grad_norm": 2.413079261779785, | |
| "learning_rate": 3.3390460526315786e-06, | |
| "loss": 0.0821, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.275, | |
| "eval_loss": 0.2597045302391052, | |
| "eval_runtime": 532.1724, | |
| "eval_samples_per_second": 3.178, | |
| "eval_steps_per_second": 0.398, | |
| "eval_wer": 10.247283626826526, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.27625, | |
| "grad_norm": 2.8475470542907715, | |
| "learning_rate": 3.3332894736842105e-06, | |
| "loss": 0.1317, | |
| "step": 5525 | |
| }, | |
| { | |
| "epoch": 0.2775, | |
| "grad_norm": 2.919682025909424, | |
| "learning_rate": 3.3275328947368416e-06, | |
| "loss": 0.1323, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 0.27875, | |
| "grad_norm": 3.0585904121398926, | |
| "learning_rate": 3.3217763157894735e-06, | |
| "loss": 0.1332, | |
| "step": 5575 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 2.4418559074401855, | |
| "learning_rate": 3.3160197368421046e-06, | |
| "loss": 0.1126, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.28125, | |
| "grad_norm": 2.9454727172851562, | |
| "learning_rate": 3.3102631578947365e-06, | |
| "loss": 0.0991, | |
| "step": 5625 | |
| }, | |
| { | |
| "epoch": 0.2825, | |
| "grad_norm": 2.472628593444824, | |
| "learning_rate": 3.304506578947368e-06, | |
| "loss": 0.1106, | |
| "step": 5650 | |
| }, | |
| { | |
| "epoch": 0.28375, | |
| "grad_norm": 2.1178548336029053, | |
| "learning_rate": 3.2987499999999995e-06, | |
| "loss": 0.1027, | |
| "step": 5675 | |
| }, | |
| { | |
| "epoch": 0.285, | |
| "grad_norm": 2.5170726776123047, | |
| "learning_rate": 3.2929934210526315e-06, | |
| "loss": 0.1027, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 0.28625, | |
| "grad_norm": 2.9180397987365723, | |
| "learning_rate": 3.287236842105263e-06, | |
| "loss": 0.1045, | |
| "step": 5725 | |
| }, | |
| { | |
| "epoch": 0.2875, | |
| "grad_norm": 2.6896932125091553, | |
| "learning_rate": 3.2814802631578945e-06, | |
| "loss": 0.1069, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 0.28875, | |
| "grad_norm": 3.1297285556793213, | |
| "learning_rate": 3.275723684210526e-06, | |
| "loss": 0.1003, | |
| "step": 5775 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 2.4746246337890625, | |
| "learning_rate": 3.269967105263158e-06, | |
| "loss": 0.1084, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.29125, | |
| "grad_norm": 1.7318406105041504, | |
| "learning_rate": 3.264210526315789e-06, | |
| "loss": 0.0846, | |
| "step": 5825 | |
| }, | |
| { | |
| "epoch": 0.2925, | |
| "grad_norm": 2.190168857574463, | |
| "learning_rate": 3.258453947368421e-06, | |
| "loss": 0.082, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 0.29375, | |
| "grad_norm": 1.5366681814193726, | |
| "learning_rate": 3.252697368421052e-06, | |
| "loss": 0.0656, | |
| "step": 5875 | |
| }, | |
| { | |
| "epoch": 0.295, | |
| "grad_norm": 1.8261510133743286, | |
| "learning_rate": 3.246940789473684e-06, | |
| "loss": 0.0646, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 0.29625, | |
| "grad_norm": 1.9088908433914185, | |
| "learning_rate": 3.2411842105263155e-06, | |
| "loss": 0.0662, | |
| "step": 5925 | |
| }, | |
| { | |
| "epoch": 0.2975, | |
| "grad_norm": 1.3404430150985718, | |
| "learning_rate": 3.235427631578947e-06, | |
| "loss": 0.0712, | |
| "step": 5950 | |
| }, | |
| { | |
| "epoch": 0.29875, | |
| "grad_norm": 1.7546651363372803, | |
| "learning_rate": 3.2296710526315785e-06, | |
| "loss": 0.084, | |
| "step": 5975 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 1.7727612257003784, | |
| "learning_rate": 3.2239144736842104e-06, | |
| "loss": 0.0988, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.2406572848558426, | |
| "eval_runtime": 535.6321, | |
| "eval_samples_per_second": 3.157, | |
| "eval_steps_per_second": 0.396, | |
| "eval_wer": 9.447983014861997, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.30125, | |
| "grad_norm": 2.477670907974243, | |
| "learning_rate": 3.2181578947368415e-06, | |
| "loss": 0.1013, | |
| "step": 6025 | |
| }, | |
| { | |
| "epoch": 0.3025, | |
| "grad_norm": 4.175459384918213, | |
| "learning_rate": 3.2124013157894734e-06, | |
| "loss": 0.1199, | |
| "step": 6050 | |
| }, | |
| { | |
| "epoch": 0.30375, | |
| "grad_norm": 2.4588561058044434, | |
| "learning_rate": 3.2066447368421053e-06, | |
| "loss": 0.1203, | |
| "step": 6075 | |
| }, | |
| { | |
| "epoch": 0.305, | |
| "grad_norm": 3.759526491165161, | |
| "learning_rate": 3.2008881578947364e-06, | |
| "loss": 0.1261, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 0.30625, | |
| "grad_norm": 3.186166524887085, | |
| "learning_rate": 3.1951315789473683e-06, | |
| "loss": 0.0946, | |
| "step": 6125 | |
| }, | |
| { | |
| "epoch": 0.3075, | |
| "grad_norm": 1.874886155128479, | |
| "learning_rate": 3.1893749999999994e-06, | |
| "loss": 0.0707, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 0.30875, | |
| "grad_norm": 1.673767328262329, | |
| "learning_rate": 3.1836184210526314e-06, | |
| "loss": 0.0605, | |
| "step": 6175 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 2.6728780269622803, | |
| "learning_rate": 3.177861842105263e-06, | |
| "loss": 0.064, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.31125, | |
| "grad_norm": 1.245354175567627, | |
| "learning_rate": 3.1721052631578944e-06, | |
| "loss": 0.0603, | |
| "step": 6225 | |
| }, | |
| { | |
| "epoch": 0.3125, | |
| "grad_norm": 1.3173916339874268, | |
| "learning_rate": 3.166348684210526e-06, | |
| "loss": 0.067, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 0.31375, | |
| "grad_norm": 1.9218686819076538, | |
| "learning_rate": 3.160592105263158e-06, | |
| "loss": 0.0723, | |
| "step": 6275 | |
| }, | |
| { | |
| "epoch": 0.315, | |
| "grad_norm": 1.822493314743042, | |
| "learning_rate": 3.154835526315789e-06, | |
| "loss": 0.0772, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 0.31625, | |
| "grad_norm": 2.4955074787139893, | |
| "learning_rate": 3.149078947368421e-06, | |
| "loss": 0.1124, | |
| "step": 6325 | |
| }, | |
| { | |
| "epoch": 0.3175, | |
| "grad_norm": 2.448274612426758, | |
| "learning_rate": 3.1433223684210523e-06, | |
| "loss": 0.1144, | |
| "step": 6350 | |
| }, | |
| { | |
| "epoch": 0.31875, | |
| "grad_norm": 2.732297658920288, | |
| "learning_rate": 3.137565789473684e-06, | |
| "loss": 0.0983, | |
| "step": 6375 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 3.261770248413086, | |
| "learning_rate": 3.1318092105263158e-06, | |
| "loss": 0.11, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.32125, | |
| "grad_norm": 2.367335319519043, | |
| "learning_rate": 3.1260526315789473e-06, | |
| "loss": 0.1129, | |
| "step": 6425 | |
| }, | |
| { | |
| "epoch": 0.3225, | |
| "grad_norm": 2.4930291175842285, | |
| "learning_rate": 3.1202960526315788e-06, | |
| "loss": 0.1106, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 0.32375, | |
| "grad_norm": 1.8275959491729736, | |
| "learning_rate": 3.1145394736842103e-06, | |
| "loss": 0.0814, | |
| "step": 6475 | |
| }, | |
| { | |
| "epoch": 0.325, | |
| "grad_norm": 3.6453261375427246, | |
| "learning_rate": 3.1087828947368418e-06, | |
| "loss": 0.0824, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.325, | |
| "eval_loss": 0.24250419437885284, | |
| "eval_runtime": 531.4087, | |
| "eval_samples_per_second": 3.182, | |
| "eval_steps_per_second": 0.399, | |
| "eval_wer": 9.223179717746971, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.32625, | |
| "grad_norm": 2.3996527194976807, | |
| "learning_rate": 3.1030263157894733e-06, | |
| "loss": 0.0913, | |
| "step": 6525 | |
| }, | |
| { | |
| "epoch": 0.3275, | |
| "grad_norm": 3.106403350830078, | |
| "learning_rate": 3.0972697368421052e-06, | |
| "loss": 0.0969, | |
| "step": 6550 | |
| }, | |
| { | |
| "epoch": 0.32875, | |
| "grad_norm": 3.741685628890991, | |
| "learning_rate": 3.0915131578947363e-06, | |
| "loss": 0.1091, | |
| "step": 6575 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 1.6008243560791016, | |
| "learning_rate": 3.0859868421052626e-06, | |
| "loss": 0.0984, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.33125, | |
| "grad_norm": 2.268734931945801, | |
| "learning_rate": 3.0802302631578945e-06, | |
| "loss": 0.0968, | |
| "step": 6625 | |
| }, | |
| { | |
| "epoch": 0.3325, | |
| "grad_norm": 2.442617654800415, | |
| "learning_rate": 3.074473684210526e-06, | |
| "loss": 0.0716, | |
| "step": 6650 | |
| }, | |
| { | |
| "epoch": 0.33375, | |
| "grad_norm": 1.9763257503509521, | |
| "learning_rate": 3.0687171052631575e-06, | |
| "loss": 0.0674, | |
| "step": 6675 | |
| }, | |
| { | |
| "epoch": 0.335, | |
| "grad_norm": 1.828474998474121, | |
| "learning_rate": 3.0629605263157894e-06, | |
| "loss": 0.0654, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 0.33625, | |
| "grad_norm": 1.5649821758270264, | |
| "learning_rate": 3.0572039473684205e-06, | |
| "loss": 0.057, | |
| "step": 6725 | |
| }, | |
| { | |
| "epoch": 0.3375, | |
| "grad_norm": 1.911927580833435, | |
| "learning_rate": 3.0514473684210525e-06, | |
| "loss": 0.0532, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 0.33875, | |
| "grad_norm": 1.3287229537963867, | |
| "learning_rate": 3.045690789473684e-06, | |
| "loss": 0.0623, | |
| "step": 6775 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 1.7754572629928589, | |
| "learning_rate": 3.0399342105263155e-06, | |
| "loss": 0.0635, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.34125, | |
| "grad_norm": 1.9900065660476685, | |
| "learning_rate": 3.034177631578947e-06, | |
| "loss": 0.0678, | |
| "step": 6825 | |
| }, | |
| { | |
| "epoch": 0.3425, | |
| "grad_norm": 1.714850664138794, | |
| "learning_rate": 3.028421052631579e-06, | |
| "loss": 0.0654, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 0.34375, | |
| "grad_norm": 1.6401875019073486, | |
| "learning_rate": 3.02266447368421e-06, | |
| "loss": 0.0662, | |
| "step": 6875 | |
| }, | |
| { | |
| "epoch": 0.345, | |
| "grad_norm": 1.0171102285385132, | |
| "learning_rate": 3.016907894736842e-06, | |
| "loss": 0.0573, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 0.34625, | |
| "grad_norm": 1.4662336111068726, | |
| "learning_rate": 3.0111513157894734e-06, | |
| "loss": 0.0556, | |
| "step": 6925 | |
| }, | |
| { | |
| "epoch": 0.3475, | |
| "grad_norm": 1.7531720399856567, | |
| "learning_rate": 3.005394736842105e-06, | |
| "loss": 0.0501, | |
| "step": 6950 | |
| }, | |
| { | |
| "epoch": 0.34875, | |
| "grad_norm": 2.6019067764282227, | |
| "learning_rate": 2.9996381578947364e-06, | |
| "loss": 0.0629, | |
| "step": 6975 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 2.0052170753479004, | |
| "learning_rate": 2.9938815789473684e-06, | |
| "loss": 0.0678, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_loss": 0.23009631037712097, | |
| "eval_runtime": 530.6679, | |
| "eval_samples_per_second": 3.187, | |
| "eval_steps_per_second": 0.399, | |
| "eval_wer": 9.13575621331335, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.35125, | |
| "grad_norm": 2.00034761428833, | |
| "learning_rate": 2.988125e-06, | |
| "loss": 0.0582, | |
| "step": 7025 | |
| }, | |
| { | |
| "epoch": 0.3525, | |
| "grad_norm": 1.7806837558746338, | |
| "learning_rate": 2.9823684210526314e-06, | |
| "loss": 0.058, | |
| "step": 7050 | |
| }, | |
| { | |
| "epoch": 0.35375, | |
| "grad_norm": 1.4306073188781738, | |
| "learning_rate": 2.976611842105263e-06, | |
| "loss": 0.0555, | |
| "step": 7075 | |
| }, | |
| { | |
| "epoch": 0.355, | |
| "grad_norm": 1.8648333549499512, | |
| "learning_rate": 2.9708552631578944e-06, | |
| "loss": 0.0662, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 0.35625, | |
| "grad_norm": 2.046255350112915, | |
| "learning_rate": 2.9650986842105263e-06, | |
| "loss": 0.0873, | |
| "step": 7125 | |
| }, | |
| { | |
| "epoch": 0.3575, | |
| "grad_norm": 1.928809404373169, | |
| "learning_rate": 2.9593421052631574e-06, | |
| "loss": 0.0948, | |
| "step": 7150 | |
| }, | |
| { | |
| "epoch": 0.35875, | |
| "grad_norm": 2.6892471313476562, | |
| "learning_rate": 2.9535855263157893e-06, | |
| "loss": 0.1043, | |
| "step": 7175 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 1.9739983081817627, | |
| "learning_rate": 2.947828947368421e-06, | |
| "loss": 0.1037, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.36125, | |
| "grad_norm": 3.5157880783081055, | |
| "learning_rate": 2.9420723684210523e-06, | |
| "loss": 0.1139, | |
| "step": 7225 | |
| }, | |
| { | |
| "epoch": 0.3625, | |
| "grad_norm": 2.140559673309326, | |
| "learning_rate": 2.936315789473684e-06, | |
| "loss": 0.0912, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 0.36375, | |
| "grad_norm": 2.24043607711792, | |
| "learning_rate": 2.9305592105263158e-06, | |
| "loss": 0.0555, | |
| "step": 7275 | |
| }, | |
| { | |
| "epoch": 0.365, | |
| "grad_norm": 1.5429259538650513, | |
| "learning_rate": 2.924802631578947e-06, | |
| "loss": 0.059, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 0.36625, | |
| "grad_norm": 1.9133890867233276, | |
| "learning_rate": 2.919046052631579e-06, | |
| "loss": 0.0576, | |
| "step": 7325 | |
| }, | |
| { | |
| "epoch": 0.3675, | |
| "grad_norm": 1.585777759552002, | |
| "learning_rate": 2.9132894736842103e-06, | |
| "loss": 0.0497, | |
| "step": 7350 | |
| }, | |
| { | |
| "epoch": 0.36875, | |
| "grad_norm": 1.5571388006210327, | |
| "learning_rate": 2.907532894736842e-06, | |
| "loss": 0.0604, | |
| "step": 7375 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 1.2344049215316772, | |
| "learning_rate": 2.9017763157894737e-06, | |
| "loss": 0.0621, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.37125, | |
| "grad_norm": 1.7708073854446411, | |
| "learning_rate": 2.896019736842105e-06, | |
| "loss": 0.0713, | |
| "step": 7425 | |
| }, | |
| { | |
| "epoch": 0.3725, | |
| "grad_norm": 2.126579999923706, | |
| "learning_rate": 2.8902631578947367e-06, | |
| "loss": 0.0661, | |
| "step": 7450 | |
| }, | |
| { | |
| "epoch": 0.37375, | |
| "grad_norm": 1.9544090032577515, | |
| "learning_rate": 2.8845065789473682e-06, | |
| "loss": 0.0626, | |
| "step": 7475 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 2.478142499923706, | |
| "learning_rate": 2.8787499999999998e-06, | |
| "loss": 0.1124, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "eval_loss": 0.2558789849281311, | |
| "eval_runtime": 531.6323, | |
| "eval_samples_per_second": 3.181, | |
| "eval_steps_per_second": 0.399, | |
| "eval_wer": 9.323092294242539, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.37625, | |
| "grad_norm": 2.722101926803589, | |
| "learning_rate": 2.8729934210526313e-06, | |
| "loss": 0.1143, | |
| "step": 7525 | |
| }, | |
| { | |
| "epoch": 0.3775, | |
| "grad_norm": 2.3424594402313232, | |
| "learning_rate": 2.867236842105263e-06, | |
| "loss": 0.0968, | |
| "step": 7550 | |
| }, | |
| { | |
| "epoch": 0.37875, | |
| "grad_norm": 2.566340208053589, | |
| "learning_rate": 2.8614802631578943e-06, | |
| "loss": 0.0932, | |
| "step": 7575 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 3.2237472534179688, | |
| "learning_rate": 2.855723684210526e-06, | |
| "loss": 0.1155, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.38125, | |
| "grad_norm": 3.058669090270996, | |
| "learning_rate": 2.8499671052631573e-06, | |
| "loss": 0.103, | |
| "step": 7625 | |
| }, | |
| { | |
| "epoch": 0.3825, | |
| "grad_norm": 4.729414463043213, | |
| "learning_rate": 2.8442105263157892e-06, | |
| "loss": 0.0922, | |
| "step": 7650 | |
| }, | |
| { | |
| "epoch": 0.38375, | |
| "grad_norm": 2.140126943588257, | |
| "learning_rate": 2.8384539473684207e-06, | |
| "loss": 0.0982, | |
| "step": 7675 | |
| }, | |
| { | |
| "epoch": 0.385, | |
| "grad_norm": 2.778568983078003, | |
| "learning_rate": 2.8326973684210522e-06, | |
| "loss": 0.0993, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 0.38625, | |
| "grad_norm": 2.6681206226348877, | |
| "learning_rate": 2.826940789473684e-06, | |
| "loss": 0.1018, | |
| "step": 7725 | |
| }, | |
| { | |
| "epoch": 0.3875, | |
| "grad_norm": 1.5673187971115112, | |
| "learning_rate": 2.8211842105263157e-06, | |
| "loss": 0.0854, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 0.38875, | |
| "grad_norm": 1.3890910148620605, | |
| "learning_rate": 2.815427631578947e-06, | |
| "loss": 0.0703, | |
| "step": 7775 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 2.176023483276367, | |
| "learning_rate": 2.8096710526315787e-06, | |
| "loss": 0.0672, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 0.39125, | |
| "grad_norm": 1.2905758619308472, | |
| "learning_rate": 2.8039144736842106e-06, | |
| "loss": 0.0615, | |
| "step": 7825 | |
| }, | |
| { | |
| "epoch": 0.3925, | |
| "grad_norm": 1.3446353673934937, | |
| "learning_rate": 2.7981578947368417e-06, | |
| "loss": 0.0637, | |
| "step": 7850 | |
| }, | |
| { | |
| "epoch": 0.39375, | |
| "grad_norm": 2.1519501209259033, | |
| "learning_rate": 2.7924013157894736e-06, | |
| "loss": 0.056, | |
| "step": 7875 | |
| }, | |
| { | |
| "epoch": 0.395, | |
| "grad_norm": 1.8618980646133423, | |
| "learning_rate": 2.7866447368421047e-06, | |
| "loss": 0.0573, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 0.39625, | |
| "grad_norm": 2.5565106868743896, | |
| "learning_rate": 2.7808881578947366e-06, | |
| "loss": 0.0882, | |
| "step": 7925 | |
| }, | |
| { | |
| "epoch": 0.3975, | |
| "grad_norm": 3.98923397064209, | |
| "learning_rate": 2.775131578947368e-06, | |
| "loss": 0.0981, | |
| "step": 7950 | |
| }, | |
| { | |
| "epoch": 0.39875, | |
| "grad_norm": 3.326756477355957, | |
| "learning_rate": 2.7693749999999996e-06, | |
| "loss": 0.147, | |
| "step": 7975 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 2.8089091777801514, | |
| "learning_rate": 2.763618421052631e-06, | |
| "loss": 0.1122, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.22397179901599884, | |
| "eval_runtime": 531.6557, | |
| "eval_samples_per_second": 3.181, | |
| "eval_steps_per_second": 0.399, | |
| "eval_wer": 8.523791682278006, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.40125, | |
| "grad_norm": 1.8123100996017456, | |
| "learning_rate": 2.757861842105263e-06, | |
| "loss": 0.0967, | |
| "step": 8025 | |
| }, | |
| { | |
| "epoch": 0.4025, | |
| "grad_norm": 2.1731700897216797, | |
| "learning_rate": 2.7521052631578946e-06, | |
| "loss": 0.0927, | |
| "step": 8050 | |
| }, | |
| { | |
| "epoch": 0.40375, | |
| "grad_norm": 2.9888458251953125, | |
| "learning_rate": 2.746348684210526e-06, | |
| "loss": 0.1061, | |
| "step": 8075 | |
| }, | |
| { | |
| "epoch": 0.405, | |
| "grad_norm": 3.51106595993042, | |
| "learning_rate": 2.740592105263158e-06, | |
| "loss": 0.158, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 0.40625, | |
| "grad_norm": 3.410916805267334, | |
| "learning_rate": 2.734835526315789e-06, | |
| "loss": 0.1011, | |
| "step": 8125 | |
| }, | |
| { | |
| "epoch": 0.4075, | |
| "grad_norm": 2.426023006439209, | |
| "learning_rate": 2.729078947368421e-06, | |
| "loss": 0.0864, | |
| "step": 8150 | |
| }, | |
| { | |
| "epoch": 0.40875, | |
| "grad_norm": 2.8296170234680176, | |
| "learning_rate": 2.723322368421052e-06, | |
| "loss": 0.0929, | |
| "step": 8175 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 2.028474807739258, | |
| "learning_rate": 2.717565789473684e-06, | |
| "loss": 0.0848, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 0.41125, | |
| "grad_norm": 2.4663166999816895, | |
| "learning_rate": 2.7118092105263155e-06, | |
| "loss": 0.0698, | |
| "step": 8225 | |
| }, | |
| { | |
| "epoch": 0.4125, | |
| "grad_norm": 1.7618118524551392, | |
| "learning_rate": 2.706052631578947e-06, | |
| "loss": 0.058, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 0.41375, | |
| "grad_norm": 2.2708559036254883, | |
| "learning_rate": 2.7002960526315786e-06, | |
| "loss": 0.0607, | |
| "step": 8275 | |
| }, | |
| { | |
| "epoch": 0.415, | |
| "grad_norm": 1.6543164253234863, | |
| "learning_rate": 2.6945394736842105e-06, | |
| "loss": 0.0556, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 0.41625, | |
| "grad_norm": 2.5951287746429443, | |
| "learning_rate": 2.6887828947368416e-06, | |
| "loss": 0.0576, | |
| "step": 8325 | |
| }, | |
| { | |
| "epoch": 0.4175, | |
| "grad_norm": 1.1910465955734253, | |
| "learning_rate": 2.6830263157894735e-06, | |
| "loss": 0.059, | |
| "step": 8350 | |
| }, | |
| { | |
| "epoch": 0.41875, | |
| "grad_norm": 1.667228102684021, | |
| "learning_rate": 2.6772697368421054e-06, | |
| "loss": 0.0521, | |
| "step": 8375 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 2.1288628578186035, | |
| "learning_rate": 2.6715131578947365e-06, | |
| "loss": 0.0557, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 0.42125, | |
| "grad_norm": 2.0485122203826904, | |
| "learning_rate": 2.6657565789473684e-06, | |
| "loss": 0.0493, | |
| "step": 8425 | |
| }, | |
| { | |
| "epoch": 0.4225, | |
| "grad_norm": 1.8512142896652222, | |
| "learning_rate": 2.6599999999999995e-06, | |
| "loss": 0.056, | |
| "step": 8450 | |
| }, | |
| { | |
| "epoch": 0.42375, | |
| "grad_norm": 1.8958942890167236, | |
| "learning_rate": 2.6542434210526314e-06, | |
| "loss": 0.059, | |
| "step": 8475 | |
| }, | |
| { | |
| "epoch": 0.425, | |
| "grad_norm": 1.2833645343780518, | |
| "learning_rate": 2.648486842105263e-06, | |
| "loss": 0.0477, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.425, | |
| "eval_loss": 0.23789365589618683, | |
| "eval_runtime": 530.7286, | |
| "eval_samples_per_second": 3.186, | |
| "eval_steps_per_second": 0.399, | |
| "eval_wer": 8.317721993255901, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.42625, | |
| "grad_norm": 1.2612178325653076, | |
| "learning_rate": 2.6427302631578945e-06, | |
| "loss": 0.05, | |
| "step": 8525 | |
| }, | |
| { | |
| "epoch": 0.4275, | |
| "grad_norm": 2.100247621536255, | |
| "learning_rate": 2.636973684210526e-06, | |
| "loss": 0.0626, | |
| "step": 8550 | |
| }, | |
| { | |
| "epoch": 0.42875, | |
| "grad_norm": 2.7199559211730957, | |
| "learning_rate": 2.631217105263158e-06, | |
| "loss": 0.0906, | |
| "step": 8575 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 3.267314910888672, | |
| "learning_rate": 2.625460526315789e-06, | |
| "loss": 0.1068, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 0.43125, | |
| "grad_norm": 3.2623515129089355, | |
| "learning_rate": 2.619703947368421e-06, | |
| "loss": 0.0849, | |
| "step": 8625 | |
| }, | |
| { | |
| "epoch": 0.4325, | |
| "grad_norm": 1.8294329643249512, | |
| "learning_rate": 2.613947368421052e-06, | |
| "loss": 0.0776, | |
| "step": 8650 | |
| }, | |
| { | |
| "epoch": 0.43375, | |
| "grad_norm": 3.3888967037200928, | |
| "learning_rate": 2.608190789473684e-06, | |
| "loss": 0.0869, | |
| "step": 8675 | |
| }, | |
| { | |
| "epoch": 0.435, | |
| "grad_norm": 2.5059332847595215, | |
| "learning_rate": 2.602434210526316e-06, | |
| "loss": 0.0781, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 0.43625, | |
| "grad_norm": 1.8527718782424927, | |
| "learning_rate": 2.596677631578947e-06, | |
| "loss": 0.0513, | |
| "step": 8725 | |
| }, | |
| { | |
| "epoch": 0.4375, | |
| "grad_norm": 1.4375104904174805, | |
| "learning_rate": 2.590921052631579e-06, | |
| "loss": 0.053, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 0.43875, | |
| "grad_norm": 1.923519253730774, | |
| "learning_rate": 2.5851644736842104e-06, | |
| "loss": 0.0487, | |
| "step": 8775 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 1.6237260103225708, | |
| "learning_rate": 2.579407894736842e-06, | |
| "loss": 0.0499, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 0.44125, | |
| "grad_norm": 1.7452889680862427, | |
| "learning_rate": 2.5736513157894734e-06, | |
| "loss": 0.0538, | |
| "step": 8825 | |
| }, | |
| { | |
| "epoch": 0.4425, | |
| "grad_norm": 1.7012261152267456, | |
| "learning_rate": 2.5678947368421053e-06, | |
| "loss": 0.0529, | |
| "step": 8850 | |
| }, | |
| { | |
| "epoch": 0.44375, | |
| "grad_norm": 1.8288905620574951, | |
| "learning_rate": 2.5621381578947364e-06, | |
| "loss": 0.0473, | |
| "step": 8875 | |
| }, | |
| { | |
| "epoch": 0.445, | |
| "grad_norm": 1.9288239479064941, | |
| "learning_rate": 2.5563815789473683e-06, | |
| "loss": 0.0683, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 0.44625, | |
| "grad_norm": 1.3186031579971313, | |
| "learning_rate": 2.5506249999999994e-06, | |
| "loss": 0.065, | |
| "step": 8925 | |
| }, | |
| { | |
| "epoch": 0.4475, | |
| "grad_norm": 1.340890645980835, | |
| "learning_rate": 2.5448684210526313e-06, | |
| "loss": 0.0617, | |
| "step": 8950 | |
| }, | |
| { | |
| "epoch": 0.44875, | |
| "grad_norm": 2.7007381916046143, | |
| "learning_rate": 2.539111842105263e-06, | |
| "loss": 0.0512, | |
| "step": 8975 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 1.688952922821045, | |
| "learning_rate": 2.5333552631578943e-06, | |
| "loss": 0.0638, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_loss": 0.23539182543754578, | |
| "eval_runtime": 531.3204, | |
| "eval_samples_per_second": 3.183, | |
| "eval_steps_per_second": 0.399, | |
| "eval_wer": 8.948420132384163, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.45125, | |
| "grad_norm": 3.06502103805542, | |
| "learning_rate": 2.527598684210526e-06, | |
| "loss": 0.0691, | |
| "step": 9025 | |
| }, | |
| { | |
| "epoch": 0.4525, | |
| "grad_norm": 2.2699365615844727, | |
| "learning_rate": 2.5218421052631578e-06, | |
| "loss": 0.0765, | |
| "step": 9050 | |
| }, | |
| { | |
| "epoch": 0.45375, | |
| "grad_norm": 3.057246208190918, | |
| "learning_rate": 2.5160855263157893e-06, | |
| "loss": 0.0874, | |
| "step": 9075 | |
| }, | |
| { | |
| "epoch": 0.455, | |
| "grad_norm": 2.452810764312744, | |
| "learning_rate": 2.5105592105263156e-06, | |
| "loss": 0.0992, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 0.45625, | |
| "grad_norm": 1.8321553468704224, | |
| "learning_rate": 2.504802631578947e-06, | |
| "loss": 0.091, | |
| "step": 9125 | |
| }, | |
| { | |
| "epoch": 0.4575, | |
| "grad_norm": 2.1675491333007812, | |
| "learning_rate": 2.499046052631579e-06, | |
| "loss": 0.0968, | |
| "step": 9150 | |
| }, | |
| { | |
| "epoch": 0.45875, | |
| "grad_norm": 2.440648317337036, | |
| "learning_rate": 2.49328947368421e-06, | |
| "loss": 0.094, | |
| "step": 9175 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 3.3630011081695557, | |
| "learning_rate": 2.487532894736842e-06, | |
| "loss": 0.0934, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 0.46125, | |
| "grad_norm": 3.1267924308776855, | |
| "learning_rate": 2.481776315789473e-06, | |
| "loss": 0.095, | |
| "step": 9225 | |
| }, | |
| { | |
| "epoch": 0.4625, | |
| "grad_norm": 2.791846752166748, | |
| "learning_rate": 2.476019736842105e-06, | |
| "loss": 0.0988, | |
| "step": 9250 | |
| }, | |
| { | |
| "epoch": 0.46375, | |
| "grad_norm": 1.883380651473999, | |
| "learning_rate": 2.4702631578947365e-06, | |
| "loss": 0.089, | |
| "step": 9275 | |
| }, | |
| { | |
| "epoch": 0.465, | |
| "grad_norm": 2.572441577911377, | |
| "learning_rate": 2.464506578947368e-06, | |
| "loss": 0.0933, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 0.46625, | |
| "grad_norm": 3.08231258392334, | |
| "learning_rate": 2.45875e-06, | |
| "loss": 0.0856, | |
| "step": 9325 | |
| }, | |
| { | |
| "epoch": 0.4675, | |
| "grad_norm": 2.208491563796997, | |
| "learning_rate": 2.4529934210526315e-06, | |
| "loss": 0.0795, | |
| "step": 9350 | |
| }, | |
| { | |
| "epoch": 0.46875, | |
| "grad_norm": 2.896657943725586, | |
| "learning_rate": 2.447236842105263e-06, | |
| "loss": 0.0625, | |
| "step": 9375 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 1.3465672731399536, | |
| "learning_rate": 2.4414802631578945e-06, | |
| "loss": 0.0822, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 0.47125, | |
| "grad_norm": 3.4039506912231445, | |
| "learning_rate": 2.4357236842105264e-06, | |
| "loss": 0.0813, | |
| "step": 9425 | |
| }, | |
| { | |
| "epoch": 0.4725, | |
| "grad_norm": 2.213761568069458, | |
| "learning_rate": 2.4299671052631575e-06, | |
| "loss": 0.075, | |
| "step": 9450 | |
| }, | |
| { | |
| "epoch": 0.47375, | |
| "grad_norm": 1.693393588066101, | |
| "learning_rate": 2.4242105263157894e-06, | |
| "loss": 0.0819, | |
| "step": 9475 | |
| }, | |
| { | |
| "epoch": 0.475, | |
| "grad_norm": 3.1261212825775146, | |
| "learning_rate": 2.4184539473684205e-06, | |
| "loss": 0.0735, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.475, | |
| "eval_loss": 0.22311098873615265, | |
| "eval_runtime": 530.3307, | |
| "eval_samples_per_second": 3.189, | |
| "eval_steps_per_second": 0.4, | |
| "eval_wer": 8.39890096165855, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.47625, | |
| "grad_norm": 2.6011083126068115, | |
| "learning_rate": 2.4126973684210524e-06, | |
| "loss": 0.0557, | |
| "step": 9525 | |
| }, | |
| { | |
| "epoch": 0.4775, | |
| "grad_norm": 1.4606833457946777, | |
| "learning_rate": 2.406940789473684e-06, | |
| "loss": 0.0518, | |
| "step": 9550 | |
| }, | |
| { | |
| "epoch": 0.47875, | |
| "grad_norm": 3.201547145843506, | |
| "learning_rate": 2.4011842105263154e-06, | |
| "loss": 0.0616, | |
| "step": 9575 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.470755696296692, | |
| "learning_rate": 2.395427631578947e-06, | |
| "loss": 0.0566, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 0.48125, | |
| "grad_norm": 1.0501068830490112, | |
| "learning_rate": 2.389671052631579e-06, | |
| "loss": 0.0482, | |
| "step": 9625 | |
| }, | |
| { | |
| "epoch": 0.4825, | |
| "grad_norm": 1.7576944828033447, | |
| "learning_rate": 2.38391447368421e-06, | |
| "loss": 0.0487, | |
| "step": 9650 | |
| }, | |
| { | |
| "epoch": 0.48375, | |
| "grad_norm": 2.6596386432647705, | |
| "learning_rate": 2.378157894736842e-06, | |
| "loss": 0.0548, | |
| "step": 9675 | |
| }, | |
| { | |
| "epoch": 0.485, | |
| "grad_norm": 2.2998361587524414, | |
| "learning_rate": 2.372401315789474e-06, | |
| "loss": 0.0755, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 0.48625, | |
| "grad_norm": 1.885953426361084, | |
| "learning_rate": 2.366644736842105e-06, | |
| "loss": 0.0635, | |
| "step": 9725 | |
| }, | |
| { | |
| "epoch": 0.4875, | |
| "grad_norm": 1.686090111732483, | |
| "learning_rate": 2.360888157894737e-06, | |
| "loss": 0.0664, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 0.48875, | |
| "grad_norm": 1.487586259841919, | |
| "learning_rate": 2.3551315789473683e-06, | |
| "loss": 0.0723, | |
| "step": 9775 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 1.5484004020690918, | |
| "learning_rate": 2.349375e-06, | |
| "loss": 0.0697, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 0.49125, | |
| "grad_norm": 1.6730592250823975, | |
| "learning_rate": 2.3436184210526314e-06, | |
| "loss": 0.0726, | |
| "step": 9825 | |
| }, | |
| { | |
| "epoch": 0.4925, | |
| "grad_norm": 1.549166202545166, | |
| "learning_rate": 2.337861842105263e-06, | |
| "loss": 0.0599, | |
| "step": 9850 | |
| }, | |
| { | |
| "epoch": 0.49375, | |
| "grad_norm": 2.127182960510254, | |
| "learning_rate": 2.3321052631578944e-06, | |
| "loss": 0.0552, | |
| "step": 9875 | |
| }, | |
| { | |
| "epoch": 0.495, | |
| "grad_norm": 1.5453063249588013, | |
| "learning_rate": 2.3263486842105263e-06, | |
| "loss": 0.0647, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 0.49625, | |
| "grad_norm": 2.2514312267303467, | |
| "learning_rate": 2.3205921052631574e-06, | |
| "loss": 0.0543, | |
| "step": 9925 | |
| }, | |
| { | |
| "epoch": 0.4975, | |
| "grad_norm": 1.5466394424438477, | |
| "learning_rate": 2.3148355263157893e-06, | |
| "loss": 0.0576, | |
| "step": 9950 | |
| }, | |
| { | |
| "epoch": 0.49875, | |
| "grad_norm": 1.1446313858032227, | |
| "learning_rate": 2.309078947368421e-06, | |
| "loss": 0.0581, | |
| "step": 9975 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.7810652256011963, | |
| "learning_rate": 2.3033223684210523e-06, | |
| "loss": 0.0548, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.23302312195301056, | |
| "eval_runtime": 531.2942, | |
| "eval_samples_per_second": 3.183, | |
| "eval_steps_per_second": 0.399, | |
| "eval_wer": 8.57374797052579, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.50125, | |
| "grad_norm": 1.8410784006118774, | |
| "learning_rate": 2.2975657894736842e-06, | |
| "loss": 0.0818, | |
| "step": 10025 | |
| }, | |
| { | |
| "epoch": 0.5025, | |
| "grad_norm": 2.0660974979400635, | |
| "learning_rate": 2.2918092105263158e-06, | |
| "loss": 0.0839, | |
| "step": 10050 | |
| }, | |
| { | |
| "epoch": 0.50375, | |
| "grad_norm": 0.7626898288726807, | |
| "learning_rate": 2.2860526315789473e-06, | |
| "loss": 0.0717, | |
| "step": 10075 | |
| }, | |
| { | |
| "epoch": 0.505, | |
| "grad_norm": 3.819746971130371, | |
| "learning_rate": 2.2802960526315788e-06, | |
| "loss": 0.0696, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 0.50625, | |
| "grad_norm": 1.8556462526321411, | |
| "learning_rate": 2.2745394736842103e-06, | |
| "loss": 0.0776, | |
| "step": 10125 | |
| }, | |
| { | |
| "epoch": 0.5075, | |
| "grad_norm": 2.1852500438690186, | |
| "learning_rate": 2.2687828947368418e-06, | |
| "loss": 0.1228, | |
| "step": 10150 | |
| }, | |
| { | |
| "epoch": 0.50875, | |
| "grad_norm": 2.6284213066101074, | |
| "learning_rate": 2.2630263157894737e-06, | |
| "loss": 0.1121, | |
| "step": 10175 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 3.840794801712036, | |
| "learning_rate": 2.2572697368421048e-06, | |
| "loss": 0.087, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 0.51125, | |
| "grad_norm": 1.920469045639038, | |
| "learning_rate": 2.2515131578947367e-06, | |
| "loss": 0.0876, | |
| "step": 10225 | |
| }, | |
| { | |
| "epoch": 0.5125, | |
| "grad_norm": 2.9199891090393066, | |
| "learning_rate": 2.2457565789473682e-06, | |
| "loss": 0.0812, | |
| "step": 10250 | |
| }, | |
| { | |
| "epoch": 0.51375, | |
| "grad_norm": 2.7151129245758057, | |
| "learning_rate": 2.2399999999999997e-06, | |
| "loss": 0.0733, | |
| "step": 10275 | |
| }, | |
| { | |
| "epoch": 0.515, | |
| "grad_norm": 3.474050760269165, | |
| "learning_rate": 2.2342434210526312e-06, | |
| "loss": 0.0934, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 0.51625, | |
| "grad_norm": 1.5654582977294922, | |
| "learning_rate": 2.228486842105263e-06, | |
| "loss": 0.0762, | |
| "step": 10325 | |
| }, | |
| { | |
| "epoch": 0.5175, | |
| "grad_norm": 1.0436935424804688, | |
| "learning_rate": 2.2227302631578947e-06, | |
| "loss": 0.0727, | |
| "step": 10350 | |
| }, | |
| { | |
| "epoch": 0.51875, | |
| "grad_norm": 0.8793361186981201, | |
| "learning_rate": 2.216973684210526e-06, | |
| "loss": 0.0471, | |
| "step": 10375 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.7731598019599915, | |
| "learning_rate": 2.211217105263158e-06, | |
| "loss": 0.0467, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 0.52125, | |
| "grad_norm": 1.2689337730407715, | |
| "learning_rate": 2.205460526315789e-06, | |
| "loss": 0.0485, | |
| "step": 10425 | |
| }, | |
| { | |
| "epoch": 0.5225, | |
| "grad_norm": 1.4495617151260376, | |
| "learning_rate": 2.199703947368421e-06, | |
| "loss": 0.0541, | |
| "step": 10450 | |
| }, | |
| { | |
| "epoch": 0.52375, | |
| "grad_norm": 1.4262604713439941, | |
| "learning_rate": 2.193947368421052e-06, | |
| "loss": 0.0539, | |
| "step": 10475 | |
| }, | |
| { | |
| "epoch": 0.525, | |
| "grad_norm": 1.8088651895523071, | |
| "learning_rate": 2.188190789473684e-06, | |
| "loss": 0.0557, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.525, | |
| "eval_loss": 0.2133007049560547, | |
| "eval_runtime": 530.2894, | |
| "eval_samples_per_second": 3.189, | |
| "eval_steps_per_second": 0.4, | |
| "eval_wer": 8.361433745472711, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.52625, | |
| "grad_norm": 1.178223967552185, | |
| "learning_rate": 2.1824342105263156e-06, | |
| "loss": 0.0549, | |
| "step": 10525 | |
| }, | |
| { | |
| "epoch": 0.5275, | |
| "grad_norm": 1.7510823011398315, | |
| "learning_rate": 2.176677631578947e-06, | |
| "loss": 0.0557, | |
| "step": 10550 | |
| }, | |
| { | |
| "epoch": 0.52875, | |
| "grad_norm": 0.9500125050544739, | |
| "learning_rate": 2.1709210526315786e-06, | |
| "loss": 0.0553, | |
| "step": 10575 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 2.060792922973633, | |
| "learning_rate": 2.1651644736842106e-06, | |
| "loss": 0.0596, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 0.53125, | |
| "grad_norm": 2.1061859130859375, | |
| "learning_rate": 2.159407894736842e-06, | |
| "loss": 0.0539, | |
| "step": 10625 | |
| }, | |
| { | |
| "epoch": 0.5325, | |
| "grad_norm": 1.6122857332229614, | |
| "learning_rate": 2.1536513157894736e-06, | |
| "loss": 0.053, | |
| "step": 10650 | |
| }, | |
| { | |
| "epoch": 0.53375, | |
| "grad_norm": 2.2909045219421387, | |
| "learning_rate": 2.147894736842105e-06, | |
| "loss": 0.0614, | |
| "step": 10675 | |
| }, | |
| { | |
| "epoch": 0.535, | |
| "grad_norm": 3.2241578102111816, | |
| "learning_rate": 2.1421381578947366e-06, | |
| "loss": 0.0829, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 0.53625, | |
| "grad_norm": 2.7384145259857178, | |
| "learning_rate": 2.136611842105263e-06, | |
| "loss": 0.0817, | |
| "step": 10725 | |
| }, | |
| { | |
| "epoch": 0.5375, | |
| "grad_norm": 1.8319401741027832, | |
| "learning_rate": 2.1308552631578944e-06, | |
| "loss": 0.0823, | |
| "step": 10750 | |
| }, | |
| { | |
| "epoch": 0.53875, | |
| "grad_norm": 2.4007859230041504, | |
| "learning_rate": 2.125098684210526e-06, | |
| "loss": 0.0733, | |
| "step": 10775 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 2.042520046234131, | |
| "learning_rate": 2.119342105263158e-06, | |
| "loss": 0.0838, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 0.54125, | |
| "grad_norm": 2.0478389263153076, | |
| "learning_rate": 2.1135855263157893e-06, | |
| "loss": 0.0831, | |
| "step": 10825 | |
| }, | |
| { | |
| "epoch": 0.5425, | |
| "grad_norm": 2.357926607131958, | |
| "learning_rate": 2.107828947368421e-06, | |
| "loss": 0.0728, | |
| "step": 10850 | |
| }, | |
| { | |
| "epoch": 0.54375, | |
| "grad_norm": 2.214553117752075, | |
| "learning_rate": 2.1020723684210523e-06, | |
| "loss": 0.0804, | |
| "step": 10875 | |
| }, | |
| { | |
| "epoch": 0.545, | |
| "grad_norm": 3.484598398208618, | |
| "learning_rate": 2.0963157894736843e-06, | |
| "loss": 0.0592, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 0.54625, | |
| "grad_norm": 1.5546646118164062, | |
| "learning_rate": 2.0905592105263158e-06, | |
| "loss": 0.0577, | |
| "step": 10925 | |
| }, | |
| { | |
| "epoch": 0.5475, | |
| "grad_norm": 2.218691349029541, | |
| "learning_rate": 2.0848026315789473e-06, | |
| "loss": 0.053, | |
| "step": 10950 | |
| }, | |
| { | |
| "epoch": 0.54875, | |
| "grad_norm": 2.9559834003448486, | |
| "learning_rate": 2.0790460526315788e-06, | |
| "loss": 0.0543, | |
| "step": 10975 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 1.6290831565856934, | |
| "learning_rate": 2.0732894736842103e-06, | |
| "loss": 0.0626, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_loss": 0.2083810567855835, | |
| "eval_runtime": 531.6457, | |
| "eval_samples_per_second": 3.181, | |
| "eval_steps_per_second": 0.399, | |
| "eval_wer": 8.286499313101036, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.55125, | |
| "grad_norm": 2.2507994174957275, | |
| "learning_rate": 2.067532894736842e-06, | |
| "loss": 0.0645, | |
| "step": 11025 | |
| }, | |
| { | |
| "epoch": 0.5525, | |
| "grad_norm": 3.930997133255005, | |
| "learning_rate": 2.0617763157894733e-06, | |
| "loss": 0.0699, | |
| "step": 11050 | |
| }, | |
| { | |
| "epoch": 0.55375, | |
| "grad_norm": 3.1073126792907715, | |
| "learning_rate": 2.056019736842105e-06, | |
| "loss": 0.0852, | |
| "step": 11075 | |
| }, | |
| { | |
| "epoch": 0.555, | |
| "grad_norm": 2.5678088665008545, | |
| "learning_rate": 2.0502631578947367e-06, | |
| "loss": 0.0863, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 0.55625, | |
| "grad_norm": 2.97763729095459, | |
| "learning_rate": 2.0445065789473682e-06, | |
| "loss": 0.0718, | |
| "step": 11125 | |
| }, | |
| { | |
| "epoch": 0.5575, | |
| "grad_norm": 1.2580708265304565, | |
| "learning_rate": 2.0387499999999998e-06, | |
| "loss": 0.0462, | |
| "step": 11150 | |
| }, | |
| { | |
| "epoch": 0.55875, | |
| "grad_norm": 1.804002285003662, | |
| "learning_rate": 2.0329934210526317e-06, | |
| "loss": 0.0364, | |
| "step": 11175 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 1.492600679397583, | |
| "learning_rate": 2.027236842105263e-06, | |
| "loss": 0.0438, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 0.56125, | |
| "grad_norm": 2.423004627227783, | |
| "learning_rate": 2.0214802631578947e-06, | |
| "loss": 0.031, | |
| "step": 11225 | |
| }, | |
| { | |
| "epoch": 0.5625, | |
| "grad_norm": 1.5198426246643066, | |
| "learning_rate": 2.015723684210526e-06, | |
| "loss": 0.0324, | |
| "step": 11250 | |
| }, | |
| { | |
| "epoch": 0.56375, | |
| "grad_norm": 0.9852400422096252, | |
| "learning_rate": 2.0099671052631577e-06, | |
| "loss": 0.029, | |
| "step": 11275 | |
| }, | |
| { | |
| "epoch": 0.565, | |
| "grad_norm": 1.2327955961227417, | |
| "learning_rate": 2.004210526315789e-06, | |
| "loss": 0.0406, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 0.56625, | |
| "grad_norm": 1.455636978149414, | |
| "learning_rate": 1.9984539473684207e-06, | |
| "loss": 0.047, | |
| "step": 11325 | |
| }, | |
| { | |
| "epoch": 0.5675, | |
| "grad_norm": 1.4720903635025024, | |
| "learning_rate": 1.9926973684210522e-06, | |
| "loss": 0.0444, | |
| "step": 11350 | |
| }, | |
| { | |
| "epoch": 0.56875, | |
| "grad_norm": 1.7255401611328125, | |
| "learning_rate": 1.986940789473684e-06, | |
| "loss": 0.0514, | |
| "step": 11375 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 1.3503352403640747, | |
| "learning_rate": 1.9811842105263157e-06, | |
| "loss": 0.0533, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 0.57125, | |
| "grad_norm": 1.5066325664520264, | |
| "learning_rate": 1.975427631578947e-06, | |
| "loss": 0.0524, | |
| "step": 11425 | |
| }, | |
| { | |
| "epoch": 0.5725, | |
| "grad_norm": 1.877842903137207, | |
| "learning_rate": 1.9696710526315787e-06, | |
| "loss": 0.0519, | |
| "step": 11450 | |
| }, | |
| { | |
| "epoch": 0.57375, | |
| "grad_norm": 1.4466218948364258, | |
| "learning_rate": 1.9639144736842106e-06, | |
| "loss": 0.0548, | |
| "step": 11475 | |
| }, | |
| { | |
| "epoch": 0.575, | |
| "grad_norm": 1.3053616285324097, | |
| "learning_rate": 1.958157894736842e-06, | |
| "loss": 0.0472, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.575, | |
| "eval_loss": 0.23307645320892334, | |
| "eval_runtime": 536.26, | |
| "eval_samples_per_second": 3.153, | |
| "eval_steps_per_second": 0.395, | |
| "eval_wer": 8.074185088047958, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.57625, | |
| "grad_norm": 1.172753930091858, | |
| "learning_rate": 1.9524013157894736e-06, | |
| "loss": 0.0506, | |
| "step": 11525 | |
| }, | |
| { | |
| "epoch": 0.5775, | |
| "grad_norm": 1.700363039970398, | |
| "learning_rate": 1.946644736842105e-06, | |
| "loss": 0.0585, | |
| "step": 11550 | |
| }, | |
| { | |
| "epoch": 0.57875, | |
| "grad_norm": 1.3203791379928589, | |
| "learning_rate": 1.9408881578947366e-06, | |
| "loss": 0.0499, | |
| "step": 11575 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 1.4109314680099487, | |
| "learning_rate": 1.935131578947368e-06, | |
| "loss": 0.0433, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 0.58125, | |
| "grad_norm": 1.3247355222702026, | |
| "learning_rate": 1.929375e-06, | |
| "loss": 0.0378, | |
| "step": 11625 | |
| }, | |
| { | |
| "epoch": 0.5825, | |
| "grad_norm": 0.9325533509254456, | |
| "learning_rate": 1.9236184210526316e-06, | |
| "loss": 0.0442, | |
| "step": 11650 | |
| }, | |
| { | |
| "epoch": 0.58375, | |
| "grad_norm": 1.8996745347976685, | |
| "learning_rate": 1.917861842105263e-06, | |
| "loss": 0.049, | |
| "step": 11675 | |
| }, | |
| { | |
| "epoch": 0.585, | |
| "grad_norm": 1.7976350784301758, | |
| "learning_rate": 1.9121052631578946e-06, | |
| "loss": 0.0467, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 0.58625, | |
| "grad_norm": 2.180805206298828, | |
| "learning_rate": 1.906348684210526e-06, | |
| "loss": 0.0493, | |
| "step": 11725 | |
| }, | |
| { | |
| "epoch": 0.5875, | |
| "grad_norm": 1.2519850730895996, | |
| "learning_rate": 1.9005921052631576e-06, | |
| "loss": 0.0486, | |
| "step": 11750 | |
| }, | |
| { | |
| "epoch": 0.58875, | |
| "grad_norm": 2.3758866786956787, | |
| "learning_rate": 1.8948355263157893e-06, | |
| "loss": 0.0584, | |
| "step": 11775 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 2.0312483310699463, | |
| "learning_rate": 1.8890789473684208e-06, | |
| "loss": 0.0702, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 0.59125, | |
| "grad_norm": 2.017726182937622, | |
| "learning_rate": 1.8833223684210525e-06, | |
| "loss": 0.0822, | |
| "step": 11825 | |
| }, | |
| { | |
| "epoch": 0.5925, | |
| "grad_norm": 2.159196138381958, | |
| "learning_rate": 1.8775657894736842e-06, | |
| "loss": 0.0918, | |
| "step": 11850 | |
| }, | |
| { | |
| "epoch": 0.59375, | |
| "grad_norm": 2.8051164150238037, | |
| "learning_rate": 1.8718092105263158e-06, | |
| "loss": 0.0927, | |
| "step": 11875 | |
| }, | |
| { | |
| "epoch": 0.595, | |
| "grad_norm": 1.9617701768875122, | |
| "learning_rate": 1.8660526315789473e-06, | |
| "loss": 0.0762, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 0.59625, | |
| "grad_norm": 1.4993948936462402, | |
| "learning_rate": 1.8602960526315788e-06, | |
| "loss": 0.0768, | |
| "step": 11925 | |
| }, | |
| { | |
| "epoch": 0.5975, | |
| "grad_norm": 2.1341333389282227, | |
| "learning_rate": 1.8545394736842105e-06, | |
| "loss": 0.0647, | |
| "step": 11950 | |
| }, | |
| { | |
| "epoch": 0.59875, | |
| "grad_norm": 1.5004290342330933, | |
| "learning_rate": 1.848782894736842e-06, | |
| "loss": 0.0669, | |
| "step": 11975 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 3.0987565517425537, | |
| "learning_rate": 1.8430263157894735e-06, | |
| "loss": 0.0636, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.2118152379989624, | |
| "eval_runtime": 536.0484, | |
| "eval_samples_per_second": 3.155, | |
| "eval_steps_per_second": 0.395, | |
| "eval_wer": 7.961783439490445, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.60125, | |
| "grad_norm": 1.6456586122512817, | |
| "learning_rate": 1.837269736842105e-06, | |
| "loss": 0.0701, | |
| "step": 12025 | |
| }, | |
| { | |
| "epoch": 0.6025, | |
| "grad_norm": 2.0990679264068604, | |
| "learning_rate": 1.8315131578947367e-06, | |
| "loss": 0.0573, | |
| "step": 12050 | |
| }, | |
| { | |
| "epoch": 0.60375, | |
| "grad_norm": 1.8728748559951782, | |
| "learning_rate": 1.8257565789473682e-06, | |
| "loss": 0.054, | |
| "step": 12075 | |
| }, | |
| { | |
| "epoch": 0.605, | |
| "grad_norm": 1.2849019765853882, | |
| "learning_rate": 1.8199999999999997e-06, | |
| "loss": 0.0522, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 0.60625, | |
| "grad_norm": 1.6803030967712402, | |
| "learning_rate": 1.8142434210526312e-06, | |
| "loss": 0.0492, | |
| "step": 12125 | |
| }, | |
| { | |
| "epoch": 0.6075, | |
| "grad_norm": 1.9102485179901123, | |
| "learning_rate": 1.808486842105263e-06, | |
| "loss": 0.0482, | |
| "step": 12150 | |
| }, | |
| { | |
| "epoch": 0.60875, | |
| "grad_norm": 1.1118731498718262, | |
| "learning_rate": 1.8027302631578947e-06, | |
| "loss": 0.0422, | |
| "step": 12175 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 1.1670501232147217, | |
| "learning_rate": 1.7969736842105262e-06, | |
| "loss": 0.0515, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 0.61125, | |
| "grad_norm": 2.522876739501953, | |
| "learning_rate": 1.7912171052631579e-06, | |
| "loss": 0.0412, | |
| "step": 12225 | |
| }, | |
| { | |
| "epoch": 0.6125, | |
| "grad_norm": 1.2704464197158813, | |
| "learning_rate": 1.7854605263157894e-06, | |
| "loss": 0.0508, | |
| "step": 12250 | |
| }, | |
| { | |
| "epoch": 0.61375, | |
| "grad_norm": 2.399094343185425, | |
| "learning_rate": 1.779703947368421e-06, | |
| "loss": 0.0547, | |
| "step": 12275 | |
| }, | |
| { | |
| "epoch": 0.615, | |
| "grad_norm": 2.2606582641601562, | |
| "learning_rate": 1.7739473684210524e-06, | |
| "loss": 0.0562, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 0.61625, | |
| "grad_norm": 0.5112090110778809, | |
| "learning_rate": 1.7681907894736841e-06, | |
| "loss": 0.0513, | |
| "step": 12325 | |
| }, | |
| { | |
| "epoch": 0.6175, | |
| "grad_norm": 1.1044148206710815, | |
| "learning_rate": 1.7624342105263156e-06, | |
| "loss": 0.0544, | |
| "step": 12350 | |
| }, | |
| { | |
| "epoch": 0.61875, | |
| "grad_norm": 1.2760109901428223, | |
| "learning_rate": 1.7566776315789471e-06, | |
| "loss": 0.0512, | |
| "step": 12375 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 1.3780227899551392, | |
| "learning_rate": 1.7509210526315786e-06, | |
| "loss": 0.0546, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 0.62125, | |
| "grad_norm": 1.0981767177581787, | |
| "learning_rate": 1.7451644736842104e-06, | |
| "loss": 0.041, | |
| "step": 12425 | |
| }, | |
| { | |
| "epoch": 0.6225, | |
| "grad_norm": 2.353482484817505, | |
| "learning_rate": 1.7394078947368419e-06, | |
| "loss": 0.0479, | |
| "step": 12450 | |
| }, | |
| { | |
| "epoch": 0.62375, | |
| "grad_norm": 1.3375900983810425, | |
| "learning_rate": 1.7336513157894734e-06, | |
| "loss": 0.0522, | |
| "step": 12475 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 2.1002514362335205, | |
| "learning_rate": 1.7278947368421053e-06, | |
| "loss": 0.0466, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "eval_loss": 0.21263667941093445, | |
| "eval_runtime": 535.5066, | |
| "eval_samples_per_second": 3.158, | |
| "eval_steps_per_second": 0.396, | |
| "eval_wer": 7.468465093043587, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.62625, | |
| "grad_norm": 1.5551177263259888, | |
| "learning_rate": 1.7221381578947368e-06, | |
| "loss": 0.0584, | |
| "step": 12525 | |
| }, | |
| { | |
| "epoch": 0.6275, | |
| "grad_norm": 2.234121322631836, | |
| "learning_rate": 1.7163815789473683e-06, | |
| "loss": 0.061, | |
| "step": 12550 | |
| }, | |
| { | |
| "epoch": 0.62875, | |
| "grad_norm": 2.269101619720459, | |
| "learning_rate": 1.7106249999999998e-06, | |
| "loss": 0.0607, | |
| "step": 12575 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 2.8848202228546143, | |
| "learning_rate": 1.7048684210526315e-06, | |
| "loss": 0.0675, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 0.63125, | |
| "grad_norm": 2.2159249782562256, | |
| "learning_rate": 1.699111842105263e-06, | |
| "loss": 0.0783, | |
| "step": 12625 | |
| }, | |
| { | |
| "epoch": 0.6325, | |
| "grad_norm": 1.5829565525054932, | |
| "learning_rate": 1.6933552631578946e-06, | |
| "loss": 0.0834, | |
| "step": 12650 | |
| }, | |
| { | |
| "epoch": 0.63375, | |
| "grad_norm": 1.9816817045211792, | |
| "learning_rate": 1.687598684210526e-06, | |
| "loss": 0.0727, | |
| "step": 12675 | |
| }, | |
| { | |
| "epoch": 0.635, | |
| "grad_norm": 2.8434395790100098, | |
| "learning_rate": 1.6818421052631578e-06, | |
| "loss": 0.0778, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 0.63625, | |
| "grad_norm": 2.4956297874450684, | |
| "learning_rate": 1.6760855263157893e-06, | |
| "loss": 0.0731, | |
| "step": 12725 | |
| }, | |
| { | |
| "epoch": 0.6375, | |
| "grad_norm": 1.7429981231689453, | |
| "learning_rate": 1.6703289473684208e-06, | |
| "loss": 0.0637, | |
| "step": 12750 | |
| }, | |
| { | |
| "epoch": 0.63875, | |
| "grad_norm": 2.3022801876068115, | |
| "learning_rate": 1.6645723684210525e-06, | |
| "loss": 0.0708, | |
| "step": 12775 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 1.621469497680664, | |
| "learning_rate": 1.658815789473684e-06, | |
| "loss": 0.0466, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 0.64125, | |
| "grad_norm": 1.7762545347213745, | |
| "learning_rate": 1.6530592105263155e-06, | |
| "loss": 0.0544, | |
| "step": 12825 | |
| }, | |
| { | |
| "epoch": 0.6425, | |
| "grad_norm": 1.568123698234558, | |
| "learning_rate": 1.6473026315789472e-06, | |
| "loss": 0.0457, | |
| "step": 12850 | |
| }, | |
| { | |
| "epoch": 0.64375, | |
| "grad_norm": 0.5994829535484314, | |
| "learning_rate": 1.641546052631579e-06, | |
| "loss": 0.0487, | |
| "step": 12875 | |
| }, | |
| { | |
| "epoch": 0.645, | |
| "grad_norm": 1.9480714797973633, | |
| "learning_rate": 1.6357894736842105e-06, | |
| "loss": 0.0503, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 0.64625, | |
| "grad_norm": 2.2603769302368164, | |
| "learning_rate": 1.630032894736842e-06, | |
| "loss": 0.0705, | |
| "step": 12925 | |
| }, | |
| { | |
| "epoch": 0.6475, | |
| "grad_norm": 2.2942919731140137, | |
| "learning_rate": 1.6242763157894737e-06, | |
| "loss": 0.0666, | |
| "step": 12950 | |
| }, | |
| { | |
| "epoch": 0.64875, | |
| "grad_norm": 2.819730758666992, | |
| "learning_rate": 1.6185197368421052e-06, | |
| "loss": 0.0736, | |
| "step": 12975 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 1.8207030296325684, | |
| "learning_rate": 1.6127631578947367e-06, | |
| "loss": 0.0604, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_loss": 0.21604977548122406, | |
| "eval_runtime": 534.676, | |
| "eval_samples_per_second": 3.163, | |
| "eval_steps_per_second": 0.397, | |
| "eval_wer": 7.655801173972773, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.65125, | |
| "grad_norm": 2.0296692848205566, | |
| "learning_rate": 1.6070065789473682e-06, | |
| "loss": 0.0745, | |
| "step": 13025 | |
| }, | |
| { | |
| "epoch": 0.6525, | |
| "grad_norm": 3.9246408939361572, | |
| "learning_rate": 1.60125e-06, | |
| "loss": 0.0862, | |
| "step": 13050 | |
| }, | |
| { | |
| "epoch": 0.65375, | |
| "grad_norm": 1.9909517765045166, | |
| "learning_rate": 1.5954934210526314e-06, | |
| "loss": 0.0676, | |
| "step": 13075 | |
| }, | |
| { | |
| "epoch": 0.655, | |
| "grad_norm": 2.652264356613159, | |
| "learning_rate": 1.589736842105263e-06, | |
| "loss": 0.0823, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 0.65625, | |
| "grad_norm": 2.1940698623657227, | |
| "learning_rate": 1.5839802631578944e-06, | |
| "loss": 0.0775, | |
| "step": 13125 | |
| }, | |
| { | |
| "epoch": 0.6575, | |
| "grad_norm": 3.084667444229126, | |
| "learning_rate": 1.5782236842105262e-06, | |
| "loss": 0.0779, | |
| "step": 13150 | |
| }, | |
| { | |
| "epoch": 0.65875, | |
| "grad_norm": 2.134045124053955, | |
| "learning_rate": 1.5724671052631579e-06, | |
| "loss": 0.0756, | |
| "step": 13175 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 2.4405481815338135, | |
| "learning_rate": 1.5667105263157894e-06, | |
| "loss": 0.075, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 0.66125, | |
| "grad_norm": 2.251408100128174, | |
| "learning_rate": 1.560953947368421e-06, | |
| "loss": 0.0668, | |
| "step": 13225 | |
| }, | |
| { | |
| "epoch": 0.6625, | |
| "grad_norm": 2.21307635307312, | |
| "learning_rate": 1.5551973684210526e-06, | |
| "loss": 0.076, | |
| "step": 13250 | |
| }, | |
| { | |
| "epoch": 0.66375, | |
| "grad_norm": 3.1692416667938232, | |
| "learning_rate": 1.549440789473684e-06, | |
| "loss": 0.0841, | |
| "step": 13275 | |
| }, | |
| { | |
| "epoch": 0.665, | |
| "grad_norm": 2.4879300594329834, | |
| "learning_rate": 1.5436842105263156e-06, | |
| "loss": 0.0785, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 0.66625, | |
| "grad_norm": 1.6188695430755615, | |
| "learning_rate": 1.5379276315789473e-06, | |
| "loss": 0.0698, | |
| "step": 13325 | |
| }, | |
| { | |
| "epoch": 0.6675, | |
| "grad_norm": 2.258192300796509, | |
| "learning_rate": 1.5321710526315788e-06, | |
| "loss": 0.0682, | |
| "step": 13350 | |
| }, | |
| { | |
| "epoch": 0.66875, | |
| "grad_norm": 1.7001844644546509, | |
| "learning_rate": 1.5264144736842103e-06, | |
| "loss": 0.0728, | |
| "step": 13375 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 2.0650229454040527, | |
| "learning_rate": 1.5206578947368418e-06, | |
| "loss": 0.0608, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 0.67125, | |
| "grad_norm": 1.0384840965270996, | |
| "learning_rate": 1.5149013157894736e-06, | |
| "loss": 0.0521, | |
| "step": 13425 | |
| }, | |
| { | |
| "epoch": 0.6725, | |
| "grad_norm": 1.458274483680725, | |
| "learning_rate": 1.509144736842105e-06, | |
| "loss": 0.0544, | |
| "step": 13450 | |
| }, | |
| { | |
| "epoch": 0.67375, | |
| "grad_norm": 1.678476095199585, | |
| "learning_rate": 1.5033881578947366e-06, | |
| "loss": 0.0478, | |
| "step": 13475 | |
| }, | |
| { | |
| "epoch": 0.675, | |
| "grad_norm": 2.1401052474975586, | |
| "learning_rate": 1.497631578947368e-06, | |
| "loss": 0.0544, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.675, | |
| "eval_loss": 0.21870100498199463, | |
| "eval_runtime": 534.1154, | |
| "eval_samples_per_second": 3.166, | |
| "eval_steps_per_second": 0.397, | |
| "eval_wer": 7.999250655676284, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.67625, | |
| "grad_norm": 1.387534737586975, | |
| "learning_rate": 1.491875e-06, | |
| "loss": 0.0497, | |
| "step": 13525 | |
| }, | |
| { | |
| "epoch": 0.6775, | |
| "grad_norm": 2.2233715057373047, | |
| "learning_rate": 1.4861184210526315e-06, | |
| "loss": 0.0628, | |
| "step": 13550 | |
| }, | |
| { | |
| "epoch": 0.67875, | |
| "grad_norm": 2.775345802307129, | |
| "learning_rate": 1.480361842105263e-06, | |
| "loss": 0.0883, | |
| "step": 13575 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 2.7996487617492676, | |
| "learning_rate": 1.4746052631578947e-06, | |
| "loss": 0.0895, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 0.68125, | |
| "grad_norm": 2.4933836460113525, | |
| "learning_rate": 1.4688486842105262e-06, | |
| "loss": 0.0876, | |
| "step": 13625 | |
| }, | |
| { | |
| "epoch": 0.6825, | |
| "grad_norm": 3.253474712371826, | |
| "learning_rate": 1.4630921052631578e-06, | |
| "loss": 0.0725, | |
| "step": 13650 | |
| }, | |
| { | |
| "epoch": 0.68375, | |
| "grad_norm": 2.5821990966796875, | |
| "learning_rate": 1.4573355263157893e-06, | |
| "loss": 0.088, | |
| "step": 13675 | |
| }, | |
| { | |
| "epoch": 0.685, | |
| "grad_norm": 3.219723701477051, | |
| "learning_rate": 1.451578947368421e-06, | |
| "loss": 0.079, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 0.68625, | |
| "grad_norm": 2.1482114791870117, | |
| "learning_rate": 1.4458223684210525e-06, | |
| "loss": 0.0715, | |
| "step": 13725 | |
| }, | |
| { | |
| "epoch": 0.6875, | |
| "grad_norm": 3.403439521789551, | |
| "learning_rate": 1.440065789473684e-06, | |
| "loss": 0.0731, | |
| "step": 13750 | |
| }, | |
| { | |
| "epoch": 0.68875, | |
| "grad_norm": 2.0612175464630127, | |
| "learning_rate": 1.4343092105263155e-06, | |
| "loss": 0.0669, | |
| "step": 13775 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 2.5637385845184326, | |
| "learning_rate": 1.4285526315789472e-06, | |
| "loss": 0.0766, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 0.69125, | |
| "grad_norm": 1.8747389316558838, | |
| "learning_rate": 1.4227960526315787e-06, | |
| "loss": 0.0723, | |
| "step": 13825 | |
| }, | |
| { | |
| "epoch": 0.6925, | |
| "grad_norm": 2.6436047554016113, | |
| "learning_rate": 1.4170394736842104e-06, | |
| "loss": 0.0694, | |
| "step": 13850 | |
| }, | |
| { | |
| "epoch": 0.69375, | |
| "grad_norm": 2.300952911376953, | |
| "learning_rate": 1.4112828947368422e-06, | |
| "loss": 0.0711, | |
| "step": 13875 | |
| }, | |
| { | |
| "epoch": 0.695, | |
| "grad_norm": 2.480396032333374, | |
| "learning_rate": 1.4055263157894737e-06, | |
| "loss": 0.0695, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 0.69625, | |
| "grad_norm": 3.047656536102295, | |
| "learning_rate": 1.3997697368421052e-06, | |
| "loss": 0.0827, | |
| "step": 13925 | |
| }, | |
| { | |
| "epoch": 0.6975, | |
| "grad_norm": 1.8521438837051392, | |
| "learning_rate": 1.3940131578947367e-06, | |
| "loss": 0.0799, | |
| "step": 13950 | |
| }, | |
| { | |
| "epoch": 0.69875, | |
| "grad_norm": 3.52673602104187, | |
| "learning_rate": 1.3882565789473684e-06, | |
| "loss": 0.0819, | |
| "step": 13975 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 2.5274155139923096, | |
| "learning_rate": 1.3824999999999999e-06, | |
| "loss": 0.07, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.21170927584171295, | |
| "eval_runtime": 534.7374, | |
| "eval_samples_per_second": 3.162, | |
| "eval_steps_per_second": 0.396, | |
| "eval_wer": 7.437242412888723, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.70125, | |
| "grad_norm": 3.9497313499450684, | |
| "learning_rate": 1.3767434210526314e-06, | |
| "loss": 0.0977, | |
| "step": 14025 | |
| }, | |
| { | |
| "epoch": 0.7025, | |
| "grad_norm": 5.4897284507751465, | |
| "learning_rate": 1.3709868421052631e-06, | |
| "loss": 0.1658, | |
| "step": 14050 | |
| }, | |
| { | |
| "epoch": 0.70375, | |
| "grad_norm": 3.0957064628601074, | |
| "learning_rate": 1.3652302631578946e-06, | |
| "loss": 0.1823, | |
| "step": 14075 | |
| }, | |
| { | |
| "epoch": 0.705, | |
| "grad_norm": 3.2891457080841064, | |
| "learning_rate": 1.3594736842105261e-06, | |
| "loss": 0.1777, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 0.70625, | |
| "grad_norm": 3.642838954925537, | |
| "learning_rate": 1.3537171052631576e-06, | |
| "loss": 0.177, | |
| "step": 14125 | |
| }, | |
| { | |
| "epoch": 0.7075, | |
| "grad_norm": 4.022505760192871, | |
| "learning_rate": 1.3479605263157894e-06, | |
| "loss": 0.1773, | |
| "step": 14150 | |
| }, | |
| { | |
| "epoch": 0.70875, | |
| "grad_norm": 3.632260799407959, | |
| "learning_rate": 1.3422039473684209e-06, | |
| "loss": 0.138, | |
| "step": 14175 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 1.6560989618301392, | |
| "learning_rate": 1.3364473684210526e-06, | |
| "loss": 0.1163, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 0.71125, | |
| "grad_norm": 1.4849154949188232, | |
| "learning_rate": 1.3306907894736843e-06, | |
| "loss": 0.1001, | |
| "step": 14225 | |
| }, | |
| { | |
| "epoch": 0.7125, | |
| "grad_norm": 2.3382551670074463, | |
| "learning_rate": 1.3249342105263158e-06, | |
| "loss": 0.0748, | |
| "step": 14250 | |
| }, | |
| { | |
| "epoch": 0.71375, | |
| "grad_norm": 3.0243709087371826, | |
| "learning_rate": 1.3191776315789473e-06, | |
| "loss": 0.0699, | |
| "step": 14275 | |
| }, | |
| { | |
| "epoch": 0.715, | |
| "grad_norm": 3.4510324001312256, | |
| "learning_rate": 1.3134210526315788e-06, | |
| "loss": 0.0822, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 0.71625, | |
| "grad_norm": 1.71156907081604, | |
| "learning_rate": 1.3076644736842105e-06, | |
| "loss": 0.0817, | |
| "step": 14325 | |
| }, | |
| { | |
| "epoch": 0.7175, | |
| "grad_norm": 1.4711543321609497, | |
| "learning_rate": 1.301907894736842e-06, | |
| "loss": 0.0573, | |
| "step": 14350 | |
| }, | |
| { | |
| "epoch": 0.71875, | |
| "grad_norm": 1.4108855724334717, | |
| "learning_rate": 1.2961513157894735e-06, | |
| "loss": 0.0518, | |
| "step": 14375 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 1.4882175922393799, | |
| "learning_rate": 1.290394736842105e-06, | |
| "loss": 0.0585, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 0.72125, | |
| "grad_norm": 1.6964808702468872, | |
| "learning_rate": 1.2846381578947368e-06, | |
| "loss": 0.0562, | |
| "step": 14425 | |
| }, | |
| { | |
| "epoch": 0.7225, | |
| "grad_norm": 1.7226653099060059, | |
| "learning_rate": 1.2788815789473683e-06, | |
| "loss": 0.0574, | |
| "step": 14450 | |
| }, | |
| { | |
| "epoch": 0.72375, | |
| "grad_norm": 2.7214572429656982, | |
| "learning_rate": 1.2731249999999998e-06, | |
| "loss": 0.0629, | |
| "step": 14475 | |
| }, | |
| { | |
| "epoch": 0.725, | |
| "grad_norm": 1.1752701997756958, | |
| "learning_rate": 1.2673684210526313e-06, | |
| "loss": 0.0534, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.725, | |
| "eval_loss": 0.13807399570941925, | |
| "eval_runtime": 533.693, | |
| "eval_samples_per_second": 3.168, | |
| "eval_steps_per_second": 0.397, | |
| "eval_wer": 7.04383664293743, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.72625, | |
| "grad_norm": 5.266875267028809, | |
| "learning_rate": 1.261611842105263e-06, | |
| "loss": 0.0553, | |
| "step": 14525 | |
| }, | |
| { | |
| "epoch": 0.7275, | |
| "grad_norm": 2.1979897022247314, | |
| "learning_rate": 1.2558552631578947e-06, | |
| "loss": 0.0498, | |
| "step": 14550 | |
| }, | |
| { | |
| "epoch": 0.72875, | |
| "grad_norm": 1.445584774017334, | |
| "learning_rate": 1.2500986842105262e-06, | |
| "loss": 0.0432, | |
| "step": 14575 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.985780656337738, | |
| "learning_rate": 1.244342105263158e-06, | |
| "loss": 0.0398, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 0.73125, | |
| "grad_norm": 1.4595451354980469, | |
| "learning_rate": 1.2385855263157894e-06, | |
| "loss": 0.0472, | |
| "step": 14625 | |
| }, | |
| { | |
| "epoch": 0.7325, | |
| "grad_norm": 1.6958725452423096, | |
| "learning_rate": 1.232828947368421e-06, | |
| "loss": 0.0451, | |
| "step": 14650 | |
| }, | |
| { | |
| "epoch": 0.73375, | |
| "grad_norm": 1.4922881126403809, | |
| "learning_rate": 1.2270723684210525e-06, | |
| "loss": 0.0483, | |
| "step": 14675 | |
| }, | |
| { | |
| "epoch": 0.735, | |
| "grad_norm": 2.243989944458008, | |
| "learning_rate": 1.2213157894736842e-06, | |
| "loss": 0.0691, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 0.73625, | |
| "grad_norm": 3.160104513168335, | |
| "learning_rate": 1.2155592105263157e-06, | |
| "loss": 0.0814, | |
| "step": 14725 | |
| }, | |
| { | |
| "epoch": 0.7375, | |
| "grad_norm": 2.0205318927764893, | |
| "learning_rate": 1.2098026315789472e-06, | |
| "loss": 0.0693, | |
| "step": 14750 | |
| }, | |
| { | |
| "epoch": 0.73875, | |
| "grad_norm": 1.519434928894043, | |
| "learning_rate": 1.2040460526315787e-06, | |
| "loss": 0.0589, | |
| "step": 14775 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 2.59538197517395, | |
| "learning_rate": 1.1982894736842104e-06, | |
| "loss": 0.0546, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 0.74125, | |
| "grad_norm": 2.137489080429077, | |
| "learning_rate": 1.192532894736842e-06, | |
| "loss": 0.0679, | |
| "step": 14825 | |
| }, | |
| { | |
| "epoch": 0.7425, | |
| "grad_norm": 1.5184602737426758, | |
| "learning_rate": 1.1867763157894734e-06, | |
| "loss": 0.0685, | |
| "step": 14850 | |
| }, | |
| { | |
| "epoch": 0.74375, | |
| "grad_norm": 2.101884365081787, | |
| "learning_rate": 1.1810197368421054e-06, | |
| "loss": 0.0526, | |
| "step": 14875 | |
| }, | |
| { | |
| "epoch": 0.745, | |
| "grad_norm": 1.778254508972168, | |
| "learning_rate": 1.1752631578947369e-06, | |
| "loss": 0.0463, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 0.74625, | |
| "grad_norm": 2.073361873626709, | |
| "learning_rate": 1.1695065789473684e-06, | |
| "loss": 0.0542, | |
| "step": 14925 | |
| }, | |
| { | |
| "epoch": 0.7475, | |
| "grad_norm": 2.091325283050537, | |
| "learning_rate": 1.1637499999999999e-06, | |
| "loss": 0.0456, | |
| "step": 14950 | |
| }, | |
| { | |
| "epoch": 0.74875, | |
| "grad_norm": 1.7418571710586548, | |
| "learning_rate": 1.1579934210526316e-06, | |
| "loss": 0.0435, | |
| "step": 14975 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 1.8316125869750977, | |
| "learning_rate": 1.152236842105263e-06, | |
| "loss": 0.046, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.14957565069198608, | |
| "eval_runtime": 534.2678, | |
| "eval_samples_per_second": 3.165, | |
| "eval_steps_per_second": 0.397, | |
| "eval_wer": 7.081303859123267, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.75125, | |
| "grad_norm": 2.781534433364868, | |
| "learning_rate": 1.1467105263157894e-06, | |
| "loss": 0.0728, | |
| "step": 15025 | |
| }, | |
| { | |
| "epoch": 0.7525, | |
| "grad_norm": 2.0675017833709717, | |
| "learning_rate": 1.1409539473684209e-06, | |
| "loss": 0.095, | |
| "step": 15050 | |
| }, | |
| { | |
| "epoch": 0.75375, | |
| "grad_norm": 3.430636167526245, | |
| "learning_rate": 1.1351973684210524e-06, | |
| "loss": 0.0966, | |
| "step": 15075 | |
| }, | |
| { | |
| "epoch": 0.755, | |
| "grad_norm": 3.50378680229187, | |
| "learning_rate": 1.129440789473684e-06, | |
| "loss": 0.1087, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 0.75625, | |
| "grad_norm": 2.9562337398529053, | |
| "learning_rate": 1.1236842105263156e-06, | |
| "loss": 0.1098, | |
| "step": 15125 | |
| }, | |
| { | |
| "epoch": 0.7575, | |
| "grad_norm": 2.7388198375701904, | |
| "learning_rate": 1.1179276315789471e-06, | |
| "loss": 0.1328, | |
| "step": 15150 | |
| }, | |
| { | |
| "epoch": 0.75875, | |
| "grad_norm": 3.3490402698516846, | |
| "learning_rate": 1.112171052631579e-06, | |
| "loss": 0.097, | |
| "step": 15175 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 1.3750718832015991, | |
| "learning_rate": 1.1064144736842105e-06, | |
| "loss": 0.0722, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 0.76125, | |
| "grad_norm": 1.7064391374588013, | |
| "learning_rate": 1.100657894736842e-06, | |
| "loss": 0.0588, | |
| "step": 15225 | |
| }, | |
| { | |
| "epoch": 0.7625, | |
| "grad_norm": 1.8604276180267334, | |
| "learning_rate": 1.0949013157894736e-06, | |
| "loss": 0.0557, | |
| "step": 15250 | |
| }, | |
| { | |
| "epoch": 0.76375, | |
| "grad_norm": 1.2240312099456787, | |
| "learning_rate": 1.0891447368421053e-06, | |
| "loss": 0.0438, | |
| "step": 15275 | |
| }, | |
| { | |
| "epoch": 0.765, | |
| "grad_norm": 1.5873894691467285, | |
| "learning_rate": 1.0833881578947368e-06, | |
| "loss": 0.0471, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 0.76625, | |
| "grad_norm": 1.645041823387146, | |
| "learning_rate": 1.0776315789473683e-06, | |
| "loss": 0.0586, | |
| "step": 15325 | |
| }, | |
| { | |
| "epoch": 0.7675, | |
| "grad_norm": 2.3403167724609375, | |
| "learning_rate": 1.0718749999999998e-06, | |
| "loss": 0.0698, | |
| "step": 15350 | |
| }, | |
| { | |
| "epoch": 0.76875, | |
| "grad_norm": 2.5629897117614746, | |
| "learning_rate": 1.0661184210526315e-06, | |
| "loss": 0.068, | |
| "step": 15375 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 2.1160974502563477, | |
| "learning_rate": 1.060361842105263e-06, | |
| "loss": 0.0771, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 0.77125, | |
| "grad_norm": 2.094522714614868, | |
| "learning_rate": 1.0546052631578947e-06, | |
| "loss": 0.0882, | |
| "step": 15425 | |
| }, | |
| { | |
| "epoch": 0.7725, | |
| "grad_norm": 2.3391168117523193, | |
| "learning_rate": 1.0488486842105262e-06, | |
| "loss": 0.0746, | |
| "step": 15450 | |
| }, | |
| { | |
| "epoch": 0.77375, | |
| "grad_norm": 2.208967924118042, | |
| "learning_rate": 1.0430921052631577e-06, | |
| "loss": 0.0725, | |
| "step": 15475 | |
| }, | |
| { | |
| "epoch": 0.775, | |
| "grad_norm": 2.7758445739746094, | |
| "learning_rate": 1.0373355263157895e-06, | |
| "loss": 0.066, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.775, | |
| "eval_loss": 0.1524539738893509, | |
| "eval_runtime": 533.899, | |
| "eval_samples_per_second": 3.167, | |
| "eval_steps_per_second": 0.397, | |
| "eval_wer": 7.00012489072062, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.77625, | |
| "grad_norm": 1.5453675985336304, | |
| "learning_rate": 1.031578947368421e-06, | |
| "loss": 0.0511, | |
| "step": 15525 | |
| }, | |
| { | |
| "epoch": 0.7775, | |
| "grad_norm": 2.0205094814300537, | |
| "learning_rate": 1.0258223684210525e-06, | |
| "loss": 0.05, | |
| "step": 15550 | |
| }, | |
| { | |
| "epoch": 0.77875, | |
| "grad_norm": 1.2804875373840332, | |
| "learning_rate": 1.020065789473684e-06, | |
| "loss": 0.0598, | |
| "step": 15575 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 2.22847843170166, | |
| "learning_rate": 1.0143092105263157e-06, | |
| "loss": 0.0686, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 0.78125, | |
| "grad_norm": 2.523324489593506, | |
| "learning_rate": 1.0085526315789472e-06, | |
| "loss": 0.1251, | |
| "step": 15625 | |
| }, | |
| { | |
| "epoch": 0.7825, | |
| "grad_norm": 1.8177152872085571, | |
| "learning_rate": 1.002796052631579e-06, | |
| "loss": 0.1014, | |
| "step": 15650 | |
| }, | |
| { | |
| "epoch": 0.78375, | |
| "grad_norm": 1.9223369359970093, | |
| "learning_rate": 9.970394736842104e-07, | |
| "loss": 0.0604, | |
| "step": 15675 | |
| }, | |
| { | |
| "epoch": 0.785, | |
| "grad_norm": 1.9404890537261963, | |
| "learning_rate": 9.91282894736842e-07, | |
| "loss": 0.0556, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 0.78625, | |
| "grad_norm": 1.354697823524475, | |
| "learning_rate": 9.855263157894737e-07, | |
| "loss": 0.0452, | |
| "step": 15725 | |
| }, | |
| { | |
| "epoch": 0.7875, | |
| "grad_norm": 0.9245623350143433, | |
| "learning_rate": 9.797697368421052e-07, | |
| "loss": 0.0536, | |
| "step": 15750 | |
| }, | |
| { | |
| "epoch": 0.78875, | |
| "grad_norm": 1.3286716938018799, | |
| "learning_rate": 9.740131578947369e-07, | |
| "loss": 0.0402, | |
| "step": 15775 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 2.337540626525879, | |
| "learning_rate": 9.682565789473684e-07, | |
| "loss": 0.0619, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 0.79125, | |
| "grad_norm": 1.3047797679901123, | |
| "learning_rate": 9.624999999999999e-07, | |
| "loss": 0.0582, | |
| "step": 15825 | |
| }, | |
| { | |
| "epoch": 0.7925, | |
| "grad_norm": 1.5523693561553955, | |
| "learning_rate": 9.567434210526314e-07, | |
| "loss": 0.0461, | |
| "step": 15850 | |
| }, | |
| { | |
| "epoch": 0.79375, | |
| "grad_norm": 0.8749285340309143, | |
| "learning_rate": 9.50986842105263e-07, | |
| "loss": 0.0458, | |
| "step": 15875 | |
| }, | |
| { | |
| "epoch": 0.795, | |
| "grad_norm": 1.0452526807785034, | |
| "learning_rate": 9.452302631578946e-07, | |
| "loss": 0.0419, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 0.79625, | |
| "grad_norm": 1.9379664659500122, | |
| "learning_rate": 9.394736842105263e-07, | |
| "loss": 0.0566, | |
| "step": 15925 | |
| }, | |
| { | |
| "epoch": 0.7975, | |
| "grad_norm": 1.316031575202942, | |
| "learning_rate": 9.337171052631578e-07, | |
| "loss": 0.0473, | |
| "step": 15950 | |
| }, | |
| { | |
| "epoch": 0.79875, | |
| "grad_norm": 1.216234564781189, | |
| "learning_rate": 9.279605263157895e-07, | |
| "loss": 0.0567, | |
| "step": 15975 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.7266921997070312, | |
| "learning_rate": 9.22203947368421e-07, | |
| "loss": 0.0632, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.14084434509277344, | |
| "eval_runtime": 535.4097, | |
| "eval_samples_per_second": 3.158, | |
| "eval_steps_per_second": 0.396, | |
| "eval_wer": 6.681653553141001, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.80125, | |
| "grad_norm": 1.8532096147537231, | |
| "learning_rate": 9.164473684210526e-07, | |
| "loss": 0.0579, | |
| "step": 16025 | |
| }, | |
| { | |
| "epoch": 0.8025, | |
| "grad_norm": 2.181915044784546, | |
| "learning_rate": 9.106907894736841e-07, | |
| "loss": 0.0757, | |
| "step": 16050 | |
| }, | |
| { | |
| "epoch": 0.80375, | |
| "grad_norm": 2.2596707344055176, | |
| "learning_rate": 9.049342105263157e-07, | |
| "loss": 0.0729, | |
| "step": 16075 | |
| }, | |
| { | |
| "epoch": 0.805, | |
| "grad_norm": 1.2219024896621704, | |
| "learning_rate": 8.991776315789473e-07, | |
| "loss": 0.0666, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 0.80625, | |
| "grad_norm": 1.135261058807373, | |
| "learning_rate": 8.934210526315789e-07, | |
| "loss": 0.0627, | |
| "step": 16125 | |
| }, | |
| { | |
| "epoch": 0.8075, | |
| "grad_norm": 1.6599974632263184, | |
| "learning_rate": 8.876644736842104e-07, | |
| "loss": 0.0477, | |
| "step": 16150 | |
| }, | |
| { | |
| "epoch": 0.80875, | |
| "grad_norm": 1.7189278602600098, | |
| "learning_rate": 8.81907894736842e-07, | |
| "loss": 0.049, | |
| "step": 16175 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.837539553642273, | |
| "learning_rate": 8.761513157894735e-07, | |
| "loss": 0.0489, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 0.81125, | |
| "grad_norm": 1.5122978687286377, | |
| "learning_rate": 8.703947368421051e-07, | |
| "loss": 0.0389, | |
| "step": 16225 | |
| }, | |
| { | |
| "epoch": 0.8125, | |
| "grad_norm": 1.7276921272277832, | |
| "learning_rate": 8.646381578947368e-07, | |
| "loss": 0.0449, | |
| "step": 16250 | |
| }, | |
| { | |
| "epoch": 0.81375, | |
| "grad_norm": 2.028928756713867, | |
| "learning_rate": 8.588815789473684e-07, | |
| "loss": 0.045, | |
| "step": 16275 | |
| }, | |
| { | |
| "epoch": 0.815, | |
| "grad_norm": 1.258401870727539, | |
| "learning_rate": 8.53125e-07, | |
| "loss": 0.0413, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 0.81625, | |
| "grad_norm": 1.2878379821777344, | |
| "learning_rate": 8.473684210526315e-07, | |
| "loss": 0.0454, | |
| "step": 16325 | |
| }, | |
| { | |
| "epoch": 0.8175, | |
| "grad_norm": 0.9309024810791016, | |
| "learning_rate": 8.416118421052631e-07, | |
| "loss": 0.0389, | |
| "step": 16350 | |
| }, | |
| { | |
| "epoch": 0.81875, | |
| "grad_norm": 0.6321396231651306, | |
| "learning_rate": 8.358552631578946e-07, | |
| "loss": 0.0282, | |
| "step": 16375 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 1.8799151182174683, | |
| "learning_rate": 8.300986842105262e-07, | |
| "loss": 0.0316, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 0.82125, | |
| "grad_norm": 0.642666220664978, | |
| "learning_rate": 8.243421052631577e-07, | |
| "loss": 0.0415, | |
| "step": 16425 | |
| }, | |
| { | |
| "epoch": 0.8225, | |
| "grad_norm": 1.199803352355957, | |
| "learning_rate": 8.185855263157894e-07, | |
| "loss": 0.0503, | |
| "step": 16450 | |
| }, | |
| { | |
| "epoch": 0.82375, | |
| "grad_norm": 1.517521858215332, | |
| "learning_rate": 8.128289473684211e-07, | |
| "loss": 0.0342, | |
| "step": 16475 | |
| }, | |
| { | |
| "epoch": 0.825, | |
| "grad_norm": 1.683922290802002, | |
| "learning_rate": 8.070723684210526e-07, | |
| "loss": 0.0437, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.825, | |
| "eval_loss": 0.1474502831697464, | |
| "eval_runtime": 533.549, | |
| "eval_samples_per_second": 3.169, | |
| "eval_steps_per_second": 0.397, | |
| "eval_wer": 6.594230048707381, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.82625, | |
| "grad_norm": 2.0518248081207275, | |
| "learning_rate": 8.013157894736842e-07, | |
| "loss": 0.0411, | |
| "step": 16525 | |
| }, | |
| { | |
| "epoch": 0.8275, | |
| "grad_norm": 1.139129638671875, | |
| "learning_rate": 7.955592105263157e-07, | |
| "loss": 0.0426, | |
| "step": 16550 | |
| }, | |
| { | |
| "epoch": 0.82875, | |
| "grad_norm": 0.7436901926994324, | |
| "learning_rate": 7.898026315789473e-07, | |
| "loss": 0.0413, | |
| "step": 16575 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.8292795419692993, | |
| "learning_rate": 7.840460526315789e-07, | |
| "loss": 0.0456, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 0.83125, | |
| "grad_norm": 2.60646390914917, | |
| "learning_rate": 7.782894736842105e-07, | |
| "loss": 0.0459, | |
| "step": 16625 | |
| }, | |
| { | |
| "epoch": 0.8325, | |
| "grad_norm": 2.15118408203125, | |
| "learning_rate": 7.72532894736842e-07, | |
| "loss": 0.0653, | |
| "step": 16650 | |
| }, | |
| { | |
| "epoch": 0.83375, | |
| "grad_norm": 1.8501421213150024, | |
| "learning_rate": 7.667763157894736e-07, | |
| "loss": 0.0702, | |
| "step": 16675 | |
| }, | |
| { | |
| "epoch": 0.835, | |
| "grad_norm": 1.579913854598999, | |
| "learning_rate": 7.610197368421051e-07, | |
| "loss": 0.0693, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 0.83625, | |
| "grad_norm": 2.4913477897644043, | |
| "learning_rate": 7.552631578947367e-07, | |
| "loss": 0.0874, | |
| "step": 16725 | |
| }, | |
| { | |
| "epoch": 0.8375, | |
| "grad_norm": 2.489863634109497, | |
| "learning_rate": 7.495065789473683e-07, | |
| "loss": 0.0642, | |
| "step": 16750 | |
| }, | |
| { | |
| "epoch": 0.83875, | |
| "grad_norm": 4.630337715148926, | |
| "learning_rate": 7.4375e-07, | |
| "loss": 0.0728, | |
| "step": 16775 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 1.708297848701477, | |
| "learning_rate": 7.379934210526316e-07, | |
| "loss": 0.056, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 0.84125, | |
| "grad_norm": 1.7515946626663208, | |
| "learning_rate": 7.322368421052631e-07, | |
| "loss": 0.0477, | |
| "step": 16825 | |
| }, | |
| { | |
| "epoch": 0.8425, | |
| "grad_norm": 1.6641236543655396, | |
| "learning_rate": 7.264802631578947e-07, | |
| "loss": 0.0508, | |
| "step": 16850 | |
| }, | |
| { | |
| "epoch": 0.84375, | |
| "grad_norm": 1.693472146987915, | |
| "learning_rate": 7.207236842105262e-07, | |
| "loss": 0.0457, | |
| "step": 16875 | |
| }, | |
| { | |
| "epoch": 0.845, | |
| "grad_norm": 0.845664381980896, | |
| "learning_rate": 7.149671052631578e-07, | |
| "loss": 0.0415, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 0.84625, | |
| "grad_norm": 1.8824065923690796, | |
| "learning_rate": 7.092105263157893e-07, | |
| "loss": 0.0481, | |
| "step": 16925 | |
| }, | |
| { | |
| "epoch": 0.8475, | |
| "grad_norm": 1.9034583568572998, | |
| "learning_rate": 7.03453947368421e-07, | |
| "loss": 0.0496, | |
| "step": 16950 | |
| }, | |
| { | |
| "epoch": 0.84875, | |
| "grad_norm": 2.6840953826904297, | |
| "learning_rate": 6.976973684210525e-07, | |
| "loss": 0.0574, | |
| "step": 16975 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 1.8385533094406128, | |
| "learning_rate": 6.919407894736842e-07, | |
| "loss": 0.0478, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_loss": 0.15727710723876953, | |
| "eval_runtime": 534.9573, | |
| "eval_samples_per_second": 3.161, | |
| "eval_steps_per_second": 0.396, | |
| "eval_wer": 6.794055201698514, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.85125, | |
| "grad_norm": 1.590932011604309, | |
| "learning_rate": 6.864144736842104e-07, | |
| "loss": 0.0589, | |
| "step": 17025 | |
| }, | |
| { | |
| "epoch": 0.8525, | |
| "grad_norm": 1.005034327507019, | |
| "learning_rate": 6.806578947368419e-07, | |
| "loss": 0.0554, | |
| "step": 17050 | |
| }, | |
| { | |
| "epoch": 0.85375, | |
| "grad_norm": 3.3872015476226807, | |
| "learning_rate": 6.749013157894737e-07, | |
| "loss": 0.0572, | |
| "step": 17075 | |
| }, | |
| { | |
| "epoch": 0.855, | |
| "grad_norm": 3.8093373775482178, | |
| "learning_rate": 6.691447368421053e-07, | |
| "loss": 0.1078, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 0.85625, | |
| "grad_norm": 2.587963581085205, | |
| "learning_rate": 6.633881578947368e-07, | |
| "loss": 0.1426, | |
| "step": 17125 | |
| }, | |
| { | |
| "epoch": 0.8575, | |
| "grad_norm": 3.9271957874298096, | |
| "learning_rate": 6.576315789473684e-07, | |
| "loss": 0.1496, | |
| "step": 17150 | |
| }, | |
| { | |
| "epoch": 0.85875, | |
| "grad_norm": 3.7258965969085693, | |
| "learning_rate": 6.518749999999999e-07, | |
| "loss": 0.1852, | |
| "step": 17175 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 4.298374652862549, | |
| "learning_rate": 6.461184210526315e-07, | |
| "loss": 0.2419, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 0.86125, | |
| "grad_norm": 6.419559478759766, | |
| "learning_rate": 6.403618421052631e-07, | |
| "loss": 0.225, | |
| "step": 17225 | |
| }, | |
| { | |
| "epoch": 0.8625, | |
| "grad_norm": 4.669430732727051, | |
| "learning_rate": 6.346052631578947e-07, | |
| "loss": 0.297, | |
| "step": 17250 | |
| }, | |
| { | |
| "epoch": 0.86375, | |
| "grad_norm": 4.676415920257568, | |
| "learning_rate": 6.288486842105262e-07, | |
| "loss": 0.2001, | |
| "step": 17275 | |
| }, | |
| { | |
| "epoch": 0.865, | |
| "grad_norm": 1.519974708557129, | |
| "learning_rate": 6.230921052631579e-07, | |
| "loss": 0.1029, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 0.86625, | |
| "grad_norm": 2.9553279876708984, | |
| "learning_rate": 6.173355263157894e-07, | |
| "loss": 0.0917, | |
| "step": 17325 | |
| }, | |
| { | |
| "epoch": 0.8675, | |
| "grad_norm": 1.5657232999801636, | |
| "learning_rate": 6.11578947368421e-07, | |
| "loss": 0.088, | |
| "step": 17350 | |
| }, | |
| { | |
| "epoch": 0.86875, | |
| "grad_norm": 3.1620709896087646, | |
| "learning_rate": 6.058223684210525e-07, | |
| "loss": 0.1046, | |
| "step": 17375 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 3.469240188598633, | |
| "learning_rate": 6.000657894736842e-07, | |
| "loss": 0.1004, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 0.87125, | |
| "grad_norm": 1.9016904830932617, | |
| "learning_rate": 5.943092105263158e-07, | |
| "loss": 0.0875, | |
| "step": 17425 | |
| }, | |
| { | |
| "epoch": 0.8725, | |
| "grad_norm": 3.1401467323303223, | |
| "learning_rate": 5.885526315789473e-07, | |
| "loss": 0.0593, | |
| "step": 17450 | |
| }, | |
| { | |
| "epoch": 0.87375, | |
| "grad_norm": 1.1564242839813232, | |
| "learning_rate": 5.827960526315789e-07, | |
| "loss": 0.0444, | |
| "step": 17475 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 0.9873404502868652, | |
| "learning_rate": 5.770394736842104e-07, | |
| "loss": 0.0418, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "eval_loss": 0.156468465924263, | |
| "eval_runtime": 534.8937, | |
| "eval_samples_per_second": 3.161, | |
| "eval_steps_per_second": 0.396, | |
| "eval_wer": 6.650430872986138, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.87625, | |
| "grad_norm": 1.499561071395874, | |
| "learning_rate": 5.71282894736842e-07, | |
| "loss": 0.0423, | |
| "step": 17525 | |
| }, | |
| { | |
| "epoch": 0.8775, | |
| "grad_norm": 1.0905530452728271, | |
| "learning_rate": 5.655263157894735e-07, | |
| "loss": 0.0496, | |
| "step": 17550 | |
| }, | |
| { | |
| "epoch": 0.87875, | |
| "grad_norm": 1.6048545837402344, | |
| "learning_rate": 5.597697368421053e-07, | |
| "loss": 0.0437, | |
| "step": 17575 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 1.5219619274139404, | |
| "learning_rate": 5.540131578947369e-07, | |
| "loss": 0.0676, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 0.88125, | |
| "grad_norm": 1.8919825553894043, | |
| "learning_rate": 5.482565789473684e-07, | |
| "loss": 0.0647, | |
| "step": 17625 | |
| }, | |
| { | |
| "epoch": 0.8825, | |
| "grad_norm": 2.4546618461608887, | |
| "learning_rate": 5.425e-07, | |
| "loss": 0.0625, | |
| "step": 17650 | |
| }, | |
| { | |
| "epoch": 0.88375, | |
| "grad_norm": 1.7209670543670654, | |
| "learning_rate": 5.367434210526315e-07, | |
| "loss": 0.0661, | |
| "step": 17675 | |
| }, | |
| { | |
| "epoch": 0.885, | |
| "grad_norm": 2.5535149574279785, | |
| "learning_rate": 5.309868421052631e-07, | |
| "loss": 0.0691, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 0.88625, | |
| "grad_norm": 3.5450563430786133, | |
| "learning_rate": 5.252302631578947e-07, | |
| "loss": 0.0603, | |
| "step": 17725 | |
| }, | |
| { | |
| "epoch": 0.8875, | |
| "grad_norm": 1.4123398065567017, | |
| "learning_rate": 5.194736842105262e-07, | |
| "loss": 0.0666, | |
| "step": 17750 | |
| }, | |
| { | |
| "epoch": 0.88875, | |
| "grad_norm": 1.427933931350708, | |
| "learning_rate": 5.137171052631578e-07, | |
| "loss": 0.0428, | |
| "step": 17775 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 1.3647822141647339, | |
| "learning_rate": 5.079605263157895e-07, | |
| "loss": 0.0382, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 0.89125, | |
| "grad_norm": 1.1601825952529907, | |
| "learning_rate": 5.02203947368421e-07, | |
| "loss": 0.0485, | |
| "step": 17825 | |
| }, | |
| { | |
| "epoch": 0.8925, | |
| "grad_norm": 1.2409619092941284, | |
| "learning_rate": 4.964473684210526e-07, | |
| "loss": 0.0439, | |
| "step": 17850 | |
| }, | |
| { | |
| "epoch": 0.89375, | |
| "grad_norm": 2.1224701404571533, | |
| "learning_rate": 4.906907894736842e-07, | |
| "loss": 0.0463, | |
| "step": 17875 | |
| }, | |
| { | |
| "epoch": 0.895, | |
| "grad_norm": 1.7053598165512085, | |
| "learning_rate": 4.849342105263158e-07, | |
| "loss": 0.0419, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 0.89625, | |
| "grad_norm": 1.2734942436218262, | |
| "learning_rate": 4.791776315789473e-07, | |
| "loss": 0.0385, | |
| "step": 17925 | |
| }, | |
| { | |
| "epoch": 0.8975, | |
| "grad_norm": 1.448438048362732, | |
| "learning_rate": 4.734210526315789e-07, | |
| "loss": 0.0397, | |
| "step": 17950 | |
| }, | |
| { | |
| "epoch": 0.89875, | |
| "grad_norm": 1.1724251508712769, | |
| "learning_rate": 4.6766447368421047e-07, | |
| "loss": 0.0454, | |
| "step": 17975 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 1.4745044708251953, | |
| "learning_rate": 4.6190789473684203e-07, | |
| "loss": 0.0382, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 0.15590737760066986, | |
| "eval_runtime": 533.3825, | |
| "eval_samples_per_second": 3.17, | |
| "eval_steps_per_second": 0.397, | |
| "eval_wer": 6.563007368552516, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.90125, | |
| "grad_norm": 1.3046791553497314, | |
| "learning_rate": 4.5615131578947364e-07, | |
| "loss": 0.0497, | |
| "step": 18025 | |
| }, | |
| { | |
| "epoch": 0.9025, | |
| "grad_norm": 2.052855968475342, | |
| "learning_rate": 4.5039473684210525e-07, | |
| "loss": 0.0542, | |
| "step": 18050 | |
| }, | |
| { | |
| "epoch": 0.90375, | |
| "grad_norm": 1.6691333055496216, | |
| "learning_rate": 4.446381578947368e-07, | |
| "loss": 0.0503, | |
| "step": 18075 | |
| }, | |
| { | |
| "epoch": 0.905, | |
| "grad_norm": 3.4304769039154053, | |
| "learning_rate": 4.3888157894736837e-07, | |
| "loss": 0.0572, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 0.90625, | |
| "grad_norm": 1.5289900302886963, | |
| "learning_rate": 4.33125e-07, | |
| "loss": 0.07, | |
| "step": 18125 | |
| }, | |
| { | |
| "epoch": 0.9075, | |
| "grad_norm": 2.5705385208129883, | |
| "learning_rate": 4.2736842105263154e-07, | |
| "loss": 0.0778, | |
| "step": 18150 | |
| }, | |
| { | |
| "epoch": 0.90875, | |
| "grad_norm": 2.34914493560791, | |
| "learning_rate": 4.216118421052631e-07, | |
| "loss": 0.0739, | |
| "step": 18175 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 2.6740806102752686, | |
| "learning_rate": 4.158552631578947e-07, | |
| "loss": 0.0621, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 0.91125, | |
| "grad_norm": 0.931742787361145, | |
| "learning_rate": 4.1009868421052627e-07, | |
| "loss": 0.0636, | |
| "step": 18225 | |
| }, | |
| { | |
| "epoch": 0.9125, | |
| "grad_norm": 1.7513364553451538, | |
| "learning_rate": 4.0434210526315783e-07, | |
| "loss": 0.0526, | |
| "step": 18250 | |
| }, | |
| { | |
| "epoch": 0.91375, | |
| "grad_norm": 1.3136606216430664, | |
| "learning_rate": 3.985855263157894e-07, | |
| "loss": 0.0469, | |
| "step": 18275 | |
| }, | |
| { | |
| "epoch": 0.915, | |
| "grad_norm": 1.2674484252929688, | |
| "learning_rate": 3.9282894736842105e-07, | |
| "loss": 0.0569, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 0.91625, | |
| "grad_norm": 2.0879714488983154, | |
| "learning_rate": 3.870723684210526e-07, | |
| "loss": 0.056, | |
| "step": 18325 | |
| }, | |
| { | |
| "epoch": 0.9175, | |
| "grad_norm": 1.6177654266357422, | |
| "learning_rate": 3.8131578947368417e-07, | |
| "loss": 0.0496, | |
| "step": 18350 | |
| }, | |
| { | |
| "epoch": 0.91875, | |
| "grad_norm": 1.267562985420227, | |
| "learning_rate": 3.755592105263158e-07, | |
| "loss": 0.0474, | |
| "step": 18375 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 1.9628887176513672, | |
| "learning_rate": 3.6980263157894734e-07, | |
| "loss": 0.0523, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 0.92125, | |
| "grad_norm": 2.450678586959839, | |
| "learning_rate": 3.640460526315789e-07, | |
| "loss": 0.0571, | |
| "step": 18425 | |
| }, | |
| { | |
| "epoch": 0.9225, | |
| "grad_norm": 3.2376692295074463, | |
| "learning_rate": 3.5828947368421046e-07, | |
| "loss": 0.0726, | |
| "step": 18450 | |
| }, | |
| { | |
| "epoch": 0.92375, | |
| "grad_norm": 1.718723177909851, | |
| "learning_rate": 3.5253289473684207e-07, | |
| "loss": 0.0759, | |
| "step": 18475 | |
| }, | |
| { | |
| "epoch": 0.925, | |
| "grad_norm": 1.7278677225112915, | |
| "learning_rate": 3.4677631578947363e-07, | |
| "loss": 0.0658, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.925, | |
| "eval_loss": 0.145228311419487, | |
| "eval_runtime": 537.5559, | |
| "eval_samples_per_second": 3.146, | |
| "eval_steps_per_second": 0.394, | |
| "eval_wer": 6.563007368552516, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.92625, | |
| "grad_norm": 2.314218044281006, | |
| "learning_rate": 3.410197368421052e-07, | |
| "loss": 0.083, | |
| "step": 18525 | |
| }, | |
| { | |
| "epoch": 0.9275, | |
| "grad_norm": 2.6032817363739014, | |
| "learning_rate": 3.3526315789473685e-07, | |
| "loss": 0.0796, | |
| "step": 18550 | |
| }, | |
| { | |
| "epoch": 0.92875, | |
| "grad_norm": 1.2821646928787231, | |
| "learning_rate": 3.295065789473684e-07, | |
| "loss": 0.0475, | |
| "step": 18575 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 1.2048566341400146, | |
| "learning_rate": 3.2374999999999997e-07, | |
| "loss": 0.0441, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 0.93125, | |
| "grad_norm": 2.205629348754883, | |
| "learning_rate": 3.179934210526316e-07, | |
| "loss": 0.0529, | |
| "step": 18625 | |
| }, | |
| { | |
| "epoch": 0.9325, | |
| "grad_norm": 0.948354959487915, | |
| "learning_rate": 3.1223684210526314e-07, | |
| "loss": 0.0491, | |
| "step": 18650 | |
| }, | |
| { | |
| "epoch": 0.93375, | |
| "grad_norm": 0.8600139617919922, | |
| "learning_rate": 3.064802631578947e-07, | |
| "loss": 0.0397, | |
| "step": 18675 | |
| }, | |
| { | |
| "epoch": 0.935, | |
| "grad_norm": 1.5570470094680786, | |
| "learning_rate": 3.0072368421052626e-07, | |
| "loss": 0.0479, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 0.93625, | |
| "grad_norm": 1.6737167835235596, | |
| "learning_rate": 2.9496710526315787e-07, | |
| "loss": 0.053, | |
| "step": 18725 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 1.7217572927474976, | |
| "learning_rate": 2.8921052631578943e-07, | |
| "loss": 0.0551, | |
| "step": 18750 | |
| }, | |
| { | |
| "epoch": 0.93875, | |
| "grad_norm": 2.207542896270752, | |
| "learning_rate": 2.83453947368421e-07, | |
| "loss": 0.0618, | |
| "step": 18775 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 1.6761177778244019, | |
| "learning_rate": 2.7769736842105265e-07, | |
| "loss": 0.0652, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 0.94125, | |
| "grad_norm": 2.9946813583374023, | |
| "learning_rate": 2.719407894736842e-07, | |
| "loss": 0.0738, | |
| "step": 18825 | |
| }, | |
| { | |
| "epoch": 0.9425, | |
| "grad_norm": 2.3663125038146973, | |
| "learning_rate": 2.6618421052631577e-07, | |
| "loss": 0.0629, | |
| "step": 18850 | |
| }, | |
| { | |
| "epoch": 0.94375, | |
| "grad_norm": 2.5888278484344482, | |
| "learning_rate": 2.6042763157894733e-07, | |
| "loss": 0.2007, | |
| "step": 18875 | |
| }, | |
| { | |
| "epoch": 0.945, | |
| "grad_norm": 1.5639821290969849, | |
| "learning_rate": 2.5467105263157894e-07, | |
| "loss": 0.2028, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 0.94625, | |
| "grad_norm": 1.5892317295074463, | |
| "learning_rate": 2.489144736842105e-07, | |
| "loss": 0.1062, | |
| "step": 18925 | |
| }, | |
| { | |
| "epoch": 0.9475, | |
| "grad_norm": 2.5038766860961914, | |
| "learning_rate": 2.431578947368421e-07, | |
| "loss": 0.06, | |
| "step": 18950 | |
| }, | |
| { | |
| "epoch": 0.94875, | |
| "grad_norm": 2.531886577606201, | |
| "learning_rate": 2.3740131578947364e-07, | |
| "loss": 0.0553, | |
| "step": 18975 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 2.5510354042053223, | |
| "learning_rate": 2.3164473684210526e-07, | |
| "loss": 0.0531, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 0.15760228037834167, | |
| "eval_runtime": 534.6809, | |
| "eval_samples_per_second": 3.163, | |
| "eval_steps_per_second": 0.396, | |
| "eval_wer": 6.662919945048083, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.95125, | |
| "grad_norm": 3.058936357498169, | |
| "learning_rate": 2.2588815789473684e-07, | |
| "loss": 0.061, | |
| "step": 19025 | |
| }, | |
| { | |
| "epoch": 0.9525, | |
| "grad_norm": 4.343925476074219, | |
| "learning_rate": 2.201315789473684e-07, | |
| "loss": 0.0775, | |
| "step": 19050 | |
| }, | |
| { | |
| "epoch": 0.95375, | |
| "grad_norm": 3.271355628967285, | |
| "learning_rate": 2.1437499999999999e-07, | |
| "loss": 0.0762, | |
| "step": 19075 | |
| }, | |
| { | |
| "epoch": 0.955, | |
| "grad_norm": 1.7924737930297852, | |
| "learning_rate": 2.0861842105263154e-07, | |
| "loss": 0.0733, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 0.95625, | |
| "grad_norm": 2.034940719604492, | |
| "learning_rate": 2.0286184210526313e-07, | |
| "loss": 0.0747, | |
| "step": 19125 | |
| }, | |
| { | |
| "epoch": 0.9575, | |
| "grad_norm": 3.0561563968658447, | |
| "learning_rate": 1.9733552631578946e-07, | |
| "loss": 0.0841, | |
| "step": 19150 | |
| }, | |
| { | |
| "epoch": 0.95875, | |
| "grad_norm": 1.5333133935928345, | |
| "learning_rate": 1.9157894736842102e-07, | |
| "loss": 0.0656, | |
| "step": 19175 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.5307198762893677, | |
| "learning_rate": 1.858223684210526e-07, | |
| "loss": 0.0532, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 0.96125, | |
| "grad_norm": 1.5663795471191406, | |
| "learning_rate": 1.8006578947368422e-07, | |
| "loss": 0.0485, | |
| "step": 19225 | |
| }, | |
| { | |
| "epoch": 0.9625, | |
| "grad_norm": 1.8204154968261719, | |
| "learning_rate": 1.7430921052631578e-07, | |
| "loss": 0.0506, | |
| "step": 19250 | |
| }, | |
| { | |
| "epoch": 0.96375, | |
| "grad_norm": 0.6307218074798584, | |
| "learning_rate": 1.6855263157894736e-07, | |
| "loss": 0.0412, | |
| "step": 19275 | |
| }, | |
| { | |
| "epoch": 0.965, | |
| "grad_norm": 1.2638368606567383, | |
| "learning_rate": 1.6279605263157892e-07, | |
| "loss": 0.0367, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 0.96625, | |
| "grad_norm": 1.448020100593567, | |
| "learning_rate": 1.570394736842105e-07, | |
| "loss": 0.0461, | |
| "step": 19325 | |
| }, | |
| { | |
| "epoch": 0.9675, | |
| "grad_norm": 1.148501992225647, | |
| "learning_rate": 1.5128289473684207e-07, | |
| "loss": 0.0321, | |
| "step": 19350 | |
| }, | |
| { | |
| "epoch": 0.96875, | |
| "grad_norm": 1.24919593334198, | |
| "learning_rate": 1.4552631578947368e-07, | |
| "loss": 0.0442, | |
| "step": 19375 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 1.437836766242981, | |
| "learning_rate": 1.3976973684210526e-07, | |
| "loss": 0.0418, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 0.97125, | |
| "grad_norm": 1.7176451683044434, | |
| "learning_rate": 1.3401315789473682e-07, | |
| "loss": 0.0423, | |
| "step": 19425 | |
| }, | |
| { | |
| "epoch": 0.9725, | |
| "grad_norm": 0.6188969016075134, | |
| "learning_rate": 1.282565789473684e-07, | |
| "loss": 0.0372, | |
| "step": 19450 | |
| }, | |
| { | |
| "epoch": 0.97375, | |
| "grad_norm": 0.7245228886604309, | |
| "learning_rate": 1.225e-07, | |
| "loss": 0.0447, | |
| "step": 19475 | |
| }, | |
| { | |
| "epoch": 0.975, | |
| "grad_norm": 1.1836830377578735, | |
| "learning_rate": 1.1674342105263156e-07, | |
| "loss": 0.0416, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.975, | |
| "eval_loss": 0.1550171822309494, | |
| "eval_runtime": 533.135, | |
| "eval_samples_per_second": 3.172, | |
| "eval_steps_per_second": 0.398, | |
| "eval_wer": 6.544273760459599, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.97625, | |
| "grad_norm": 2.7617335319519043, | |
| "learning_rate": 1.1098684210526315e-07, | |
| "loss": 0.0481, | |
| "step": 19525 | |
| }, | |
| { | |
| "epoch": 0.9775, | |
| "grad_norm": 1.8646786212921143, | |
| "learning_rate": 1.0523026315789472e-07, | |
| "loss": 0.0479, | |
| "step": 19550 | |
| }, | |
| { | |
| "epoch": 0.97875, | |
| "grad_norm": 3.3118820190429688, | |
| "learning_rate": 9.947368421052632e-08, | |
| "loss": 0.0622, | |
| "step": 19575 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 1.3400448560714722, | |
| "learning_rate": 9.371710526315789e-08, | |
| "loss": 0.0727, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 0.98125, | |
| "grad_norm": 3.044895648956299, | |
| "learning_rate": 8.796052631578946e-08, | |
| "loss": 0.0613, | |
| "step": 19625 | |
| }, | |
| { | |
| "epoch": 0.9825, | |
| "grad_norm": 3.217283248901367, | |
| "learning_rate": 8.220394736842105e-08, | |
| "loss": 0.0814, | |
| "step": 19650 | |
| }, | |
| { | |
| "epoch": 0.98375, | |
| "grad_norm": 1.3824083805084229, | |
| "learning_rate": 7.644736842105262e-08, | |
| "loss": 0.0598, | |
| "step": 19675 | |
| }, | |
| { | |
| "epoch": 0.985, | |
| "grad_norm": 1.3852965831756592, | |
| "learning_rate": 7.069078947368419e-08, | |
| "loss": 0.0504, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 0.98625, | |
| "grad_norm": 2.5811800956726074, | |
| "learning_rate": 6.493421052631578e-08, | |
| "loss": 0.0643, | |
| "step": 19725 | |
| }, | |
| { | |
| "epoch": 0.9875, | |
| "grad_norm": 1.2770925760269165, | |
| "learning_rate": 5.9177631578947364e-08, | |
| "loss": 0.0586, | |
| "step": 19750 | |
| }, | |
| { | |
| "epoch": 0.98875, | |
| "grad_norm": 2.8050851821899414, | |
| "learning_rate": 5.342105263157894e-08, | |
| "loss": 0.0623, | |
| "step": 19775 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 3.1270270347595215, | |
| "learning_rate": 4.766447368421052e-08, | |
| "loss": 0.1128, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 0.99125, | |
| "grad_norm": 2.787506103515625, | |
| "learning_rate": 4.1907894736842107e-08, | |
| "loss": 0.0945, | |
| "step": 19825 | |
| }, | |
| { | |
| "epoch": 0.9925, | |
| "grad_norm": 2.0053322315216064, | |
| "learning_rate": 3.615131578947368e-08, | |
| "loss": 0.0676, | |
| "step": 19850 | |
| }, | |
| { | |
| "epoch": 0.99375, | |
| "grad_norm": 2.319840669631958, | |
| "learning_rate": 3.0394736842105264e-08, | |
| "loss": 0.0434, | |
| "step": 19875 | |
| }, | |
| { | |
| "epoch": 0.995, | |
| "grad_norm": 1.3756728172302246, | |
| "learning_rate": 2.463815789473684e-08, | |
| "loss": 0.0458, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 0.99625, | |
| "grad_norm": 1.3499048948287964, | |
| "learning_rate": 1.8881578947368418e-08, | |
| "loss": 0.0426, | |
| "step": 19925 | |
| }, | |
| { | |
| "epoch": 0.9975, | |
| "grad_norm": 1.0059881210327148, | |
| "learning_rate": 1.3124999999999998e-08, | |
| "loss": 0.0543, | |
| "step": 19950 | |
| }, | |
| { | |
| "epoch": 0.99875, | |
| "grad_norm": 1.247534155845642, | |
| "learning_rate": 7.368421052631579e-09, | |
| "loss": 0.0526, | |
| "step": 19975 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.3380235433578491, | |
| "learning_rate": 1.6118421052631579e-09, | |
| "loss": 0.0435, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.15491345524787903, | |
| "eval_runtime": 532.2586, | |
| "eval_samples_per_second": 3.177, | |
| "eval_steps_per_second": 0.398, | |
| "eval_wer": 6.544273760459599, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 20000, | |
| "total_flos": 1.0871994580992e+21, | |
| "train_loss": 0.0028733723163604737, | |
| "train_runtime": 7510.0544, | |
| "train_samples_per_second": 42.61, | |
| "train_steps_per_second": 2.663 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 20000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0871994580992e+21, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |