| { | |
| "best_metric": 0.8695769309997559, | |
| "best_model_checkpoint": "./wav2vec2-large-robust-paper/checkpoint-4355", | |
| "epoch": 50.0, | |
| "global_step": 16750, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.9162847995758057, | |
| "eval_runtime": 53.7666, | |
| "eval_samples_per_second": 15.66, | |
| "eval_steps_per_second": 1.971, | |
| "eval_wer": 1.0, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 4.976729944886712e-05, | |
| "loss": 7.1369, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 3.342172145843506, | |
| "eval_runtime": 53.6439, | |
| "eval_samples_per_second": 15.696, | |
| "eval_steps_per_second": 1.976, | |
| "eval_wer": 1.0, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 4.82363747703613e-05, | |
| "loss": 3.3448, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 3.335538864135742, | |
| "eval_runtime": 53.7095, | |
| "eval_samples_per_second": 15.677, | |
| "eval_steps_per_second": 1.974, | |
| "eval_wer": 1.0, | |
| "step": 1005 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 3.3262581825256348, | |
| "eval_runtime": 53.7592, | |
| "eval_samples_per_second": 15.662, | |
| "eval_steps_per_second": 1.972, | |
| "eval_wer": 1.0, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 4.6705450091855484e-05, | |
| "loss": 3.3277, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.8927650451660156, | |
| "eval_runtime": 54.3633, | |
| "eval_samples_per_second": 15.488, | |
| "eval_steps_per_second": 1.95, | |
| "eval_wer": 1.0079421226156946, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 4.517452541334967e-05, | |
| "loss": 2.6655, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.7821804285049438, | |
| "eval_runtime": 54.4942, | |
| "eval_samples_per_second": 15.451, | |
| "eval_steps_per_second": 1.945, | |
| "eval_wer": 0.8787591601372186, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.3193254470825195, | |
| "eval_runtime": 54.3211, | |
| "eval_samples_per_second": 15.5, | |
| "eval_steps_per_second": 1.951, | |
| "eval_wer": 0.7055407984836553, | |
| "step": 2345 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "learning_rate": 4.364360073484385e-05, | |
| "loss": 1.4617, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.1407958269119263, | |
| "eval_runtime": 54.3418, | |
| "eval_samples_per_second": 15.495, | |
| "eval_steps_per_second": 1.951, | |
| "eval_wer": 0.607031781838568, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "learning_rate": 4.211267605633803e-05, | |
| "loss": 1.0805, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.0108448266983032, | |
| "eval_runtime": 54.8305, | |
| "eval_samples_per_second": 15.356, | |
| "eval_steps_per_second": 1.933, | |
| "eval_wer": 0.542240079020783, | |
| "step": 3015 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.9516933560371399, | |
| "eval_runtime": 54.4768, | |
| "eval_samples_per_second": 15.456, | |
| "eval_steps_per_second": 1.946, | |
| "eval_wer": 0.515383691285022, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 10.45, | |
| "learning_rate": 4.058175137783221e-05, | |
| "loss": 0.8759, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.9081984758377075, | |
| "eval_runtime": 54.3156, | |
| "eval_samples_per_second": 15.502, | |
| "eval_steps_per_second": 1.952, | |
| "eval_wer": 0.49020916480905535, | |
| "step": 3685 | |
| }, | |
| { | |
| "epoch": 11.94, | |
| "learning_rate": 3.905082669932639e-05, | |
| "loss": 0.7462, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.875782310962677, | |
| "eval_runtime": 54.5818, | |
| "eval_samples_per_second": 15.426, | |
| "eval_steps_per_second": 1.942, | |
| "eval_wer": 0.47056075390098373, | |
| "step": 4020 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.8695769309997559, | |
| "eval_runtime": 54.4851, | |
| "eval_samples_per_second": 15.454, | |
| "eval_steps_per_second": 1.945, | |
| "eval_wer": 0.457239344874995, | |
| "step": 4355 | |
| }, | |
| { | |
| "epoch": 13.43, | |
| "learning_rate": 3.7519902020820577e-05, | |
| "loss": 0.6429, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.8731432557106018, | |
| "eval_runtime": 55.2734, | |
| "eval_samples_per_second": 15.233, | |
| "eval_steps_per_second": 1.918, | |
| "eval_wer": 0.45348852730354927, | |
| "step": 4690 | |
| }, | |
| { | |
| "epoch": 14.93, | |
| "learning_rate": 3.598897734231476e-05, | |
| "loss": 0.5672, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.8749489784240723, | |
| "eval_runtime": 54.2876, | |
| "eval_samples_per_second": 15.51, | |
| "eval_steps_per_second": 1.953, | |
| "eval_wer": 0.450792210045784, | |
| "step": 5025 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.8752743005752563, | |
| "eval_runtime": 54.5671, | |
| "eval_samples_per_second": 15.431, | |
| "eval_steps_per_second": 1.943, | |
| "eval_wer": 0.4512460456238237, | |
| "step": 5360 | |
| }, | |
| { | |
| "epoch": 16.42, | |
| "learning_rate": 3.445805266380894e-05, | |
| "loss": 0.4959, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.9039203524589539, | |
| "eval_runtime": 54.5236, | |
| "eval_samples_per_second": 15.443, | |
| "eval_steps_per_second": 1.944, | |
| "eval_wer": 0.4486965575236595, | |
| "step": 5695 | |
| }, | |
| { | |
| "epoch": 17.91, | |
| "learning_rate": 3.2927127985303126e-05, | |
| "loss": 0.4456, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.9161030650138855, | |
| "eval_runtime": 54.3598, | |
| "eval_samples_per_second": 15.489, | |
| "eval_steps_per_second": 1.95, | |
| "eval_wer": 0.443303923008129, | |
| "step": 6030 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.9505767226219177, | |
| "eval_runtime": 54.2942, | |
| "eval_samples_per_second": 15.508, | |
| "eval_steps_per_second": 1.952, | |
| "eval_wer": 0.4429835684824539, | |
| "step": 6365 | |
| }, | |
| { | |
| "epoch": 19.4, | |
| "learning_rate": 3.13962033067973e-05, | |
| "loss": 0.392, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.9411688446998596, | |
| "eval_runtime": 54.3336, | |
| "eval_samples_per_second": 15.497, | |
| "eval_steps_per_second": 1.951, | |
| "eval_wer": 0.4438511953228239, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 20.9, | |
| "learning_rate": 2.9865278628291486e-05, | |
| "loss": 0.3594, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 0.9884207844734192, | |
| "eval_runtime": 54.3631, | |
| "eval_samples_per_second": 15.488, | |
| "eval_steps_per_second": 1.95, | |
| "eval_wer": 0.44162206174833485, | |
| "step": 7035 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 1.022162914276123, | |
| "eval_runtime": 54.656, | |
| "eval_samples_per_second": 15.405, | |
| "eval_steps_per_second": 1.939, | |
| "eval_wer": 0.45101912783480386, | |
| "step": 7370 | |
| }, | |
| { | |
| "epoch": 22.39, | |
| "learning_rate": 2.833435394978567e-05, | |
| "loss": 0.3175, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 1.0344542264938354, | |
| "eval_runtime": 54.3628, | |
| "eval_samples_per_second": 15.489, | |
| "eval_steps_per_second": 1.95, | |
| "eval_wer": 0.4439312839542427, | |
| "step": 7705 | |
| }, | |
| { | |
| "epoch": 23.88, | |
| "learning_rate": 2.6803429271279856e-05, | |
| "loss": 0.2947, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 1.084883213043213, | |
| "eval_runtime": 54.6337, | |
| "eval_samples_per_second": 15.412, | |
| "eval_steps_per_second": 1.94, | |
| "eval_wer": 0.4464807720544069, | |
| "step": 8040 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 1.0879119634628296, | |
| "eval_runtime": 54.4682, | |
| "eval_samples_per_second": 15.459, | |
| "eval_steps_per_second": 1.946, | |
| "eval_wer": 0.4472416140528852, | |
| "step": 8375 | |
| }, | |
| { | |
| "epoch": 25.37, | |
| "learning_rate": 2.527556644213105e-05, | |
| "loss": 0.2674, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 1.1071442365646362, | |
| "eval_runtime": 54.3908, | |
| "eval_samples_per_second": 15.481, | |
| "eval_steps_per_second": 1.949, | |
| "eval_wer": 0.4512460456238237, | |
| "step": 8710 | |
| }, | |
| { | |
| "epoch": 26.87, | |
| "learning_rate": 2.3744641763625232e-05, | |
| "loss": 0.2521, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 1.1146758794784546, | |
| "eval_runtime": 54.502, | |
| "eval_samples_per_second": 15.449, | |
| "eval_steps_per_second": 1.945, | |
| "eval_wer": 0.4494307033116649, | |
| "step": 9045 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 1.1426175832748413, | |
| "eval_runtime": 54.4541, | |
| "eval_samples_per_second": 15.463, | |
| "eval_steps_per_second": 1.947, | |
| "eval_wer": 0.4525408118317605, | |
| "step": 9380 | |
| }, | |
| { | |
| "epoch": 28.36, | |
| "learning_rate": 2.2216778934476424e-05, | |
| "loss": 0.2321, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_loss": 1.1591500043869019, | |
| "eval_runtime": 54.6264, | |
| "eval_samples_per_second": 15.414, | |
| "eval_steps_per_second": 1.94, | |
| "eval_wer": 0.4440380687961344, | |
| "step": 9715 | |
| }, | |
| { | |
| "epoch": 29.85, | |
| "learning_rate": 2.0685854255970608e-05, | |
| "loss": 0.2235, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 1.178154706954956, | |
| "eval_runtime": 54.3939, | |
| "eval_samples_per_second": 15.48, | |
| "eval_steps_per_second": 1.949, | |
| "eval_wer": 0.44495908805745027, | |
| "step": 10050 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_loss": 1.2050156593322754, | |
| "eval_runtime": 54.4103, | |
| "eval_samples_per_second": 15.475, | |
| "eval_steps_per_second": 1.948, | |
| "eval_wer": 0.4436643218495135, | |
| "step": 10385 | |
| }, | |
| { | |
| "epoch": 31.34, | |
| "learning_rate": 1.9154929577464788e-05, | |
| "loss": 0.2071, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 1.2223504781723022, | |
| "eval_runtime": 54.4403, | |
| "eval_samples_per_second": 15.466, | |
| "eval_steps_per_second": 1.947, | |
| "eval_wer": 0.4399535485937771, | |
| "step": 10720 | |
| }, | |
| { | |
| "epoch": 32.84, | |
| "learning_rate": 1.7624004898958974e-05, | |
| "loss": 0.1951, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_loss": 1.2269729375839233, | |
| "eval_runtime": 54.7494, | |
| "eval_samples_per_second": 15.379, | |
| "eval_steps_per_second": 1.936, | |
| "eval_wer": 0.4470680886848112, | |
| "step": 11055 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_loss": 1.246551275253296, | |
| "eval_runtime": 54.3927, | |
| "eval_samples_per_second": 15.48, | |
| "eval_steps_per_second": 1.949, | |
| "eval_wer": 0.44833615868227505, | |
| "step": 11390 | |
| }, | |
| { | |
| "epoch": 34.33, | |
| "learning_rate": 1.6093080220453154e-05, | |
| "loss": 0.1892, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_loss": 1.232478380203247, | |
| "eval_runtime": 54.41, | |
| "eval_samples_per_second": 15.475, | |
| "eval_steps_per_second": 1.948, | |
| "eval_wer": 0.44286343553532576, | |
| "step": 11725 | |
| }, | |
| { | |
| "epoch": 35.82, | |
| "learning_rate": 1.4562155541947337e-05, | |
| "loss": 0.1809, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 1.2754629850387573, | |
| "eval_runtime": 54.5117, | |
| "eval_samples_per_second": 15.446, | |
| "eval_steps_per_second": 1.945, | |
| "eval_wer": 0.44272995448296115, | |
| "step": 12060 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_loss": 1.2675007581710815, | |
| "eval_runtime": 54.594, | |
| "eval_samples_per_second": 15.423, | |
| "eval_steps_per_second": 1.942, | |
| "eval_wer": 0.4421693340630298, | |
| "step": 12395 | |
| }, | |
| { | |
| "epoch": 37.31, | |
| "learning_rate": 1.3031230863441517e-05, | |
| "loss": 0.1746, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_loss": 1.3022228479385376, | |
| "eval_runtime": 54.757, | |
| "eval_samples_per_second": 15.377, | |
| "eval_steps_per_second": 1.936, | |
| "eval_wer": 0.4417688909059359, | |
| "step": 12730 | |
| }, | |
| { | |
| "epoch": 38.81, | |
| "learning_rate": 1.15003061849357e-05, | |
| "loss": 0.1656, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_loss": 1.3178575038909912, | |
| "eval_runtime": 54.82, | |
| "eval_samples_per_second": 15.359, | |
| "eval_steps_per_second": 1.934, | |
| "eval_wer": 0.4407544349079648, | |
| "step": 13065 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 1.293350338935852, | |
| "eval_runtime": 54.6833, | |
| "eval_samples_per_second": 15.398, | |
| "eval_steps_per_second": 1.938, | |
| "eval_wer": 0.4425430810096507, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 40.3, | |
| "learning_rate": 9.972443355786895e-06, | |
| "loss": 0.1614, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_loss": 1.3303664922714233, | |
| "eval_runtime": 54.5126, | |
| "eval_samples_per_second": 15.446, | |
| "eval_steps_per_second": 1.945, | |
| "eval_wer": 0.44259647343059655, | |
| "step": 13735 | |
| }, | |
| { | |
| "epoch": 41.79, | |
| "learning_rate": 8.441518677281078e-06, | |
| "loss": 0.1564, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_loss": 1.3147845268249512, | |
| "eval_runtime": 54.5529, | |
| "eval_samples_per_second": 15.435, | |
| "eval_steps_per_second": 1.943, | |
| "eval_wer": 0.44196911248448284, | |
| "step": 14070 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_loss": 1.3266962766647339, | |
| "eval_runtime": 54.7201, | |
| "eval_samples_per_second": 15.387, | |
| "eval_steps_per_second": 1.937, | |
| "eval_wer": 0.44325053058718317, | |
| "step": 14405 | |
| }, | |
| { | |
| "epoch": 43.28, | |
| "learning_rate": 6.910593998775261e-06, | |
| "loss": 0.1546, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_loss": 1.3331360816955566, | |
| "eval_runtime": 54.6617, | |
| "eval_samples_per_second": 15.404, | |
| "eval_steps_per_second": 1.939, | |
| "eval_wer": 0.4412750110121868, | |
| "step": 14740 | |
| }, | |
| { | |
| "epoch": 44.78, | |
| "learning_rate": 5.379669320269443e-06, | |
| "loss": 0.1515, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_loss": 1.344504475593567, | |
| "eval_runtime": 54.543, | |
| "eval_samples_per_second": 15.437, | |
| "eval_steps_per_second": 1.943, | |
| "eval_wer": 0.43875221912249557, | |
| "step": 15075 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_loss": 1.352994441986084, | |
| "eval_runtime": 54.5619, | |
| "eval_samples_per_second": 15.432, | |
| "eval_steps_per_second": 1.943, | |
| "eval_wer": 0.4372438832307754, | |
| "step": 15410 | |
| }, | |
| { | |
| "epoch": 46.27, | |
| "learning_rate": 3.851806491120637e-06, | |
| "loss": 0.147, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_loss": 1.34428870677948, | |
| "eval_runtime": 54.5828, | |
| "eval_samples_per_second": 15.426, | |
| "eval_steps_per_second": 1.942, | |
| "eval_wer": 0.4384585608072934, | |
| "step": 15745 | |
| }, | |
| { | |
| "epoch": 47.76, | |
| "learning_rate": 2.3208818126148195e-06, | |
| "loss": 0.1447, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 1.3502939939498901, | |
| "eval_runtime": 54.5282, | |
| "eval_samples_per_second": 15.442, | |
| "eval_steps_per_second": 1.944, | |
| "eval_wer": 0.436856788178918, | |
| "step": 16080 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_loss": 1.358969807624817, | |
| "eval_runtime": 55.1854, | |
| "eval_samples_per_second": 15.258, | |
| "eval_steps_per_second": 1.921, | |
| "eval_wer": 0.4393261876476634, | |
| "step": 16415 | |
| }, | |
| { | |
| "epoch": 49.25, | |
| "learning_rate": 7.930189834660135e-07, | |
| "loss": 0.1437, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 1.3668413162231445, | |
| "eval_runtime": 54.6805, | |
| "eval_samples_per_second": 15.399, | |
| "eval_steps_per_second": 1.939, | |
| "eval_wer": 0.43717714270459307, | |
| "step": 16750 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 16750, | |
| "total_flos": 1.4087701795050537e+20, | |
| "train_loss": 0.8170821473989913, | |
| "train_runtime": 40557.4985, | |
| "train_samples_per_second": 4.13, | |
| "train_steps_per_second": 0.413 | |
| } | |
| ], | |
| "max_steps": 16750, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.4087701795050537e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |