| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.784097580066533, | |
| "global_step": 75000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 8.072407045009784e-06, | |
| "loss": 6.7755, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.622635355512068e-05, | |
| "loss": 4.3924, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.4380300065231574e-05, | |
| "loss": 3.9807, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 3.253424657534247e-05, | |
| "loss": 3.7959, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.068819308545336e-05, | |
| "loss": 3.6875, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.884213959556426e-05, | |
| "loss": 3.6157, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 5.699608610567515e-05, | |
| "loss": 3.5719, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 6.515003261578603e-05, | |
| "loss": 3.5222, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 7.330397912589694e-05, | |
| "loss": 3.498, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 8.145792563600784e-05, | |
| "loss": 3.4647, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_loss": 3.4406068325042725, | |
| "eval_runtime": 5633.1588, | |
| "eval_samples_per_second": 0.432, | |
| "eval_steps_per_second": 0.054, | |
| "eval_wer": 1.15852260825739, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 8.961187214611873e-05, | |
| "loss": 3.4349, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 9.776581865622962e-05, | |
| "loss": 3.4181, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 9.948523781162257e-05, | |
| "loss": 3.3879, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 9.87761989846564e-05, | |
| "loss": 3.3797, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 9.806716015769025e-05, | |
| "loss": 3.3632, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 9.735812133072407e-05, | |
| "loss": 3.3488, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 9.664908250375791e-05, | |
| "loss": 3.3378, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 9.594004367679175e-05, | |
| "loss": 3.3113, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 9.523100484982558e-05, | |
| "loss": 3.3031, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 9.452338410051335e-05, | |
| "loss": 3.2819, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_loss": 3.261105537414551, | |
| "eval_runtime": 7279.2824, | |
| "eval_samples_per_second": 0.334, | |
| "eval_steps_per_second": 0.042, | |
| "eval_wer": 1.7922252072031033, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 9.381434527354719e-05, | |
| "loss": 3.2647, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 9.310530644658103e-05, | |
| "loss": 3.2264, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 9.239626761961485e-05, | |
| "loss": 3.2075, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 9.168722879264869e-05, | |
| "loss": 3.1744, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 9.097818996568253e-05, | |
| "loss": 3.1484, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 9.026915113871636e-05, | |
| "loss": 3.1259, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 8.95601123117502e-05, | |
| "loss": 3.1157, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 8.885107348478403e-05, | |
| "loss": 3.1022, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 8.814203465781786e-05, | |
| "loss": 3.0838, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 8.74329958308517e-05, | |
| "loss": 3.0741, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_loss": 3.0150177478790283, | |
| "eval_runtime": 7749.0478, | |
| "eval_samples_per_second": 0.314, | |
| "eval_steps_per_second": 0.039, | |
| "eval_wer": 1.6811482450050355, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 8.672395700388554e-05, | |
| "loss": 3.0585, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 8.601491817691936e-05, | |
| "loss": 3.0499, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 8.530729742760714e-05, | |
| "loss": 3.0364, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 8.459825860064097e-05, | |
| "loss": 3.0311, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 8.388921977367481e-05, | |
| "loss": 3.0206, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 8.318018094670864e-05, | |
| "loss": 3.0151, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 8.247114211974248e-05, | |
| "loss": 3.015, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 8.176352137043025e-05, | |
| "loss": 3.0009, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 8.105448254346409e-05, | |
| "loss": 2.9991, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 8.034544371649792e-05, | |
| "loss": 2.9897, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "eval_loss": 2.9467122554779053, | |
| "eval_runtime": 7497.8191, | |
| "eval_samples_per_second": 0.324, | |
| "eval_steps_per_second": 0.041, | |
| "eval_wer": 1.880129997594331, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 7.963640488953175e-05, | |
| "loss": 2.9933, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 7.89273660625656e-05, | |
| "loss": 2.9868, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 7.821832723559942e-05, | |
| "loss": 2.9774, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 7.750928840863326e-05, | |
| "loss": 2.9887, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 7.68002495816671e-05, | |
| "loss": 2.9771, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 7.609121075470092e-05, | |
| "loss": 2.9693, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 7.538500808304263e-05, | |
| "loss": 2.9536, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 7.467596925607647e-05, | |
| "loss": 2.9497, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 7.39669304291103e-05, | |
| "loss": 2.9556, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 7.325930967979807e-05, | |
| "loss": 2.9472, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "eval_loss": 2.9073405265808105, | |
| "eval_runtime": 7894.0702, | |
| "eval_samples_per_second": 0.308, | |
| "eval_steps_per_second": 0.039, | |
| "eval_wer": 1.7444502680537224, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 7.25502708528319e-05, | |
| "loss": 2.9421, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 7.184123202586573e-05, | |
| "loss": 2.9412, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 7.113219319889958e-05, | |
| "loss": 2.9363, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 7.042315437193341e-05, | |
| "loss": 2.9404, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 6.971411554496725e-05, | |
| "loss": 2.9243, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 6.900507671800109e-05, | |
| "loss": 2.9242, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 6.829603789103491e-05, | |
| "loss": 2.9393, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 6.758699906406875e-05, | |
| "loss": 2.933, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 6.687796023710259e-05, | |
| "loss": 2.9206, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 6.617033948779036e-05, | |
| "loss": 2.9237, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "eval_loss": 2.881319999694824, | |
| "eval_runtime": 8126.3615, | |
| "eval_samples_per_second": 0.299, | |
| "eval_steps_per_second": 0.037, | |
| "eval_wer": 1.5856239783910557, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 6.546130066082419e-05, | |
| "loss": 2.9193, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 6.475226183385803e-05, | |
| "loss": 2.9006, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 6.404322300689187e-05, | |
| "loss": 2.9101, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 6.333418417992569e-05, | |
| "loss": 2.8993, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 6.262514535295953e-05, | |
| "loss": 2.9036, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 6.191610652599337e-05, | |
| "loss": 2.906, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 6.12070676990272e-05, | |
| "loss": 2.9025, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 6.049802887206104e-05, | |
| "loss": 2.9036, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 5.9788990045094874e-05, | |
| "loss": 2.8951, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 5.907995121812871e-05, | |
| "loss": 2.8923, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "eval_loss": 2.864828109741211, | |
| "eval_runtime": 8189.637, | |
| "eval_samples_per_second": 0.297, | |
| "eval_steps_per_second": 0.037, | |
| "eval_wer": 1.4885282519750724, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 5.8370912391162546e-05, | |
| "loss": 2.8938, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 5.766187356419638e-05, | |
| "loss": 2.8858, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 5.6954252814884144e-05, | |
| "loss": 2.8869, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 5.6245213987917976e-05, | |
| "loss": 2.883, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 5.553617516095182e-05, | |
| "loss": 2.8871, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 5.4827136333985654e-05, | |
| "loss": 2.8859, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 5.411809750701948e-05, | |
| "loss": 2.8822, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 5.341047675770725e-05, | |
| "loss": 2.8747, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "learning_rate": 5.270285600839502e-05, | |
| "loss": 2.8729, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 5.1993817181428863e-05, | |
| "loss": 2.8692, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "eval_loss": 2.8506710529327393, | |
| "eval_runtime": 8365.1128, | |
| "eval_samples_per_second": 0.291, | |
| "eval_steps_per_second": 0.036, | |
| "eval_wer": 1.3797645737346684, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "learning_rate": 5.1284778354462696e-05, | |
| "loss": 2.869, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 5.057573952749652e-05, | |
| "loss": 2.8641, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 4.986670070053036e-05, | |
| "loss": 2.8724, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 4.91576618735642e-05, | |
| "loss": 2.8617, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 4.8450041124251965e-05, | |
| "loss": 2.8586, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 4.7741002297285804e-05, | |
| "loss": 2.8655, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 4.703196347031964e-05, | |
| "loss": 2.866, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "learning_rate": 4.632292464335347e-05, | |
| "loss": 2.8702, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 4.561388581638731e-05, | |
| "loss": 2.8603, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "learning_rate": 4.4906265067075074e-05, | |
| "loss": 2.8652, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "eval_loss": 2.8389720916748047, | |
| "eval_runtime": 8250.6574, | |
| "eval_samples_per_second": 0.295, | |
| "eval_steps_per_second": 0.037, | |
| "eval_wer": 1.4632943395432523, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "learning_rate": 4.419722624010891e-05, | |
| "loss": 2.8589, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.3488187413142745e-05, | |
| "loss": 2.8625, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 4.2779148586176584e-05, | |
| "loss": 2.8415, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 4.207010975921042e-05, | |
| "loss": 2.8463, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "learning_rate": 4.136107093224425e-05, | |
| "loss": 2.8392, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "learning_rate": 4.065203210527809e-05, | |
| "loss": 2.8471, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 3.994299327831193e-05, | |
| "loss": 2.8507, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 3.923395445134575e-05, | |
| "loss": 2.8451, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 6.46, | |
| "learning_rate": 3.852491562437959e-05, | |
| "loss": 2.8463, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 3.781587679741343e-05, | |
| "loss": 2.8477, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "eval_loss": 2.831930160522461, | |
| "eval_runtime": 8120.4635, | |
| "eval_samples_per_second": 0.299, | |
| "eval_steps_per_second": 0.037, | |
| "eval_wer": 1.360951876558997, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 6.59, | |
| "learning_rate": 3.71082560481012e-05, | |
| "loss": 2.8473, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 6.65, | |
| "learning_rate": 3.639921722113503e-05, | |
| "loss": 2.8373, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 3.569017839416887e-05, | |
| "loss": 2.8431, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 3.49811395672027e-05, | |
| "loss": 2.838, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 3.427210074023653e-05, | |
| "loss": 2.845, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 3.356306191327037e-05, | |
| "loss": 2.8389, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "learning_rate": 3.285402308630421e-05, | |
| "loss": 2.8464, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 3.214640233699198e-05, | |
| "loss": 2.8313, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 3.143736351002581e-05, | |
| "loss": 2.8282, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 7.18, | |
| "learning_rate": 3.072832468305965e-05, | |
| "loss": 2.8333, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 7.18, | |
| "eval_loss": 2.822190999984741, | |
| "eval_runtime": 7683.7145, | |
| "eval_samples_per_second": 0.317, | |
| "eval_steps_per_second": 0.04, | |
| "eval_wer": 1.1936536074515367, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "learning_rate": 3.001928585609348e-05, | |
| "loss": 2.8288, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 2.931024702912732e-05, | |
| "loss": 2.8279, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 2.860262627981509e-05, | |
| "loss": 2.8291, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 7.44, | |
| "learning_rate": 2.7893587452848917e-05, | |
| "loss": 2.8336, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 2.7184548625882757e-05, | |
| "loss": 2.8292, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 7.57, | |
| "learning_rate": 2.6475509798916592e-05, | |
| "loss": 2.831, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 7.63, | |
| "learning_rate": 2.5766470971950425e-05, | |
| "loss": 2.8265, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 7.7, | |
| "learning_rate": 2.505743214498426e-05, | |
| "loss": 2.8229, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 7.76, | |
| "learning_rate": 2.4348393318018096e-05, | |
| "loss": 2.8232, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 2.364077256870586e-05, | |
| "loss": 2.8236, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "eval_loss": 2.8168787956237793, | |
| "eval_runtime": 8146.3496, | |
| "eval_samples_per_second": 0.299, | |
| "eval_steps_per_second": 0.037, | |
| "eval_wer": 1.3519813839639754, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 2.29317337417397e-05, | |
| "loss": 2.8204, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 2.2222694914773533e-05, | |
| "loss": 2.8284, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 2.151365608780737e-05, | |
| "loss": 2.8229, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 8.09, | |
| "learning_rate": 2.0804617260841205e-05, | |
| "loss": 2.8108, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "learning_rate": 2.009557843387504e-05, | |
| "loss": 2.8219, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 1.9386539606908873e-05, | |
| "loss": 2.816, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "learning_rate": 1.867891885759664e-05, | |
| "loss": 2.8139, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "learning_rate": 1.7969880030630477e-05, | |
| "loss": 2.819, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 1.7260841203664313e-05, | |
| "loss": 2.8137, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 1.655180237669815e-05, | |
| "loss": 2.8151, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "eval_loss": 2.811957597732544, | |
| "eval_runtime": 7992.8193, | |
| "eval_samples_per_second": 0.304, | |
| "eval_steps_per_second": 0.038, | |
| "eval_wer": 1.2878607017052808, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 8.54, | |
| "learning_rate": 1.5844181627385914e-05, | |
| "loss": 2.8081, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "learning_rate": 1.5135142800419752e-05, | |
| "loss": 2.8113, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 8.68, | |
| "learning_rate": 1.4426103973453586e-05, | |
| "loss": 2.8156, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "learning_rate": 1.3717065146487423e-05, | |
| "loss": 2.8218, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 8.81, | |
| "learning_rate": 1.3008026319521257e-05, | |
| "loss": 2.8164, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 1.2298987492555093e-05, | |
| "loss": 2.8072, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "learning_rate": 1.1589948665588929e-05, | |
| "loss": 2.8064, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 1.0880909838622763e-05, | |
| "loss": 2.8106, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 1.0171871011656599e-05, | |
| "loss": 2.8085, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 9.13, | |
| "learning_rate": 9.462832184690434e-06, | |
| "loss": 2.8081, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 9.13, | |
| "eval_loss": 2.808976888656616, | |
| "eval_runtime": 8181.5451, | |
| "eval_samples_per_second": 0.297, | |
| "eval_steps_per_second": 0.037, | |
| "eval_wer": 1.3355642940184824, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "learning_rate": 8.75379335772427e-06, | |
| "loss": 2.8072, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "learning_rate": 8.044754530758104e-06, | |
| "loss": 2.8064, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 9.33, | |
| "learning_rate": 7.33571570379194e-06, | |
| "loss": 2.8067, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 9.39, | |
| "learning_rate": 6.628094954479708e-06, | |
| "loss": 2.8033, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 9.46, | |
| "learning_rate": 5.919056127513543e-06, | |
| "loss": 2.8104, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 5.210017300547379e-06, | |
| "loss": 2.8062, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "learning_rate": 4.502396551235146e-06, | |
| "loss": 2.7983, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 9.65, | |
| "learning_rate": 3.793357724268981e-06, | |
| "loss": 2.7966, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "learning_rate": 3.0843188973028165e-06, | |
| "loss": 2.8003, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "learning_rate": 2.375280070336652e-06, | |
| "loss": 2.8038, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "eval_loss": 2.805771589279175, | |
| "eval_runtime": 8283.5766, | |
| "eval_samples_per_second": 0.294, | |
| "eval_steps_per_second": 0.037, | |
| "eval_wer": 1.3321451341000374, | |
| "step": 75000 | |
| } | |
| ], | |
| "max_steps": 76650, | |
| "num_train_epochs": 10, | |
| "total_flos": 7.309582504377754e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |