| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 60.0, | |
| "global_step": 7440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1.3172043010752688e-05, | |
| "loss": 13.9663, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "eval_loss": 15.575762748718262, | |
| "eval_runtime": 98.9721, | |
| "eval_samples_per_second": 18.632, | |
| "eval_steps_per_second": 2.334, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 2.6344086021505376e-05, | |
| "loss": 6.7623, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "eval_loss": 3.259247064590454, | |
| "eval_runtime": 99.4059, | |
| "eval_samples_per_second": 18.55, | |
| "eval_steps_per_second": 2.324, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 3.978494623655914e-05, | |
| "loss": 3.0242, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "eval_loss": 2.9369125366210938, | |
| "eval_runtime": 98.1481, | |
| "eval_samples_per_second": 18.788, | |
| "eval_steps_per_second": 2.354, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 5.32258064516129e-05, | |
| "loss": 2.8806, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "eval_loss": 2.8887438774108887, | |
| "eval_runtime": 97.5774, | |
| "eval_samples_per_second": 18.898, | |
| "eval_steps_per_second": 2.367, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 2.8344, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "eval_loss": 2.8125975131988525, | |
| "eval_runtime": 97.5647, | |
| "eval_samples_per_second": 18.9, | |
| "eval_steps_per_second": 2.368, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 8.010752688172043e-05, | |
| "loss": 2.7158, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "eval_loss": 2.304689884185791, | |
| "eval_runtime": 97.2319, | |
| "eval_samples_per_second": 18.965, | |
| "eval_steps_per_second": 2.376, | |
| "eval_wer": 1.003542793212754, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 9.35483870967742e-05, | |
| "loss": 1.7164, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "eval_loss": 0.7104876041412354, | |
| "eval_runtime": 96.4932, | |
| "eval_samples_per_second": 19.11, | |
| "eval_steps_per_second": 2.394, | |
| "eval_wer": 0.7594629871340668, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 0.0001, | |
| "loss": 1.0757, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "eval_loss": 0.5160450339317322, | |
| "eval_runtime": 97.694, | |
| "eval_samples_per_second": 18.875, | |
| "eval_steps_per_second": 2.365, | |
| "eval_wer": 0.6501025545403692, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 0.0001, | |
| "loss": 0.9208, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "eval_loss": 0.43188726902008057, | |
| "eval_runtime": 101.9111, | |
| "eval_samples_per_second": 18.094, | |
| "eval_steps_per_second": 2.267, | |
| "eval_wer": 0.5842811859034123, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 0.0001, | |
| "loss": 0.8012, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "eval_loss": 0.39437660574913025, | |
| "eval_runtime": 97.2792, | |
| "eval_samples_per_second": 18.956, | |
| "eval_steps_per_second": 2.375, | |
| "eval_wer": 0.5456833861644602, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 0.0001, | |
| "loss": 0.7334, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "eval_loss": 0.36806735396385193, | |
| "eval_runtime": 97.7448, | |
| "eval_samples_per_second": 18.865, | |
| "eval_steps_per_second": 2.363, | |
| "eval_wer": 0.5120268506432967, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6839, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "eval_loss": 0.35295596718788147, | |
| "eval_runtime": 104.7228, | |
| "eval_samples_per_second": 17.608, | |
| "eval_steps_per_second": 2.206, | |
| "eval_wer": 0.49822860339362296, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6599, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "eval_loss": 0.34405317902565, | |
| "eval_runtime": 100.6994, | |
| "eval_samples_per_second": 18.312, | |
| "eval_steps_per_second": 2.294, | |
| "eval_wer": 0.49561812418422524, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6101, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "eval_loss": 0.3226686716079712, | |
| "eval_runtime": 97.8943, | |
| "eval_samples_per_second": 18.837, | |
| "eval_steps_per_second": 2.36, | |
| "eval_wer": 0.4828454223382435, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5844, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "eval_loss": 0.34125906229019165, | |
| "eval_runtime": 98.6933, | |
| "eval_samples_per_second": 18.684, | |
| "eval_steps_per_second": 2.341, | |
| "eval_wer": 0.47445459630803655, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5754, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_loss": 0.33336642384529114, | |
| "eval_runtime": 99.0007, | |
| "eval_samples_per_second": 18.626, | |
| "eval_steps_per_second": 2.333, | |
| "eval_wer": 0.46140220026104795, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 13.71, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5611, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 13.71, | |
| "eval_loss": 0.30597391724586487, | |
| "eval_runtime": 101.2827, | |
| "eval_samples_per_second": 18.206, | |
| "eval_steps_per_second": 2.281, | |
| "eval_wer": 0.4672757784821928, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 14.52, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5476, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 14.52, | |
| "eval_loss": 0.3194342851638794, | |
| "eval_runtime": 105.7302, | |
| "eval_samples_per_second": 17.441, | |
| "eval_steps_per_second": 2.185, | |
| "eval_wer": 0.44881596121573747, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 15.32, | |
| "learning_rate": 0.0001, | |
| "loss": 0.526, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 15.32, | |
| "eval_loss": 0.30019786953926086, | |
| "eval_runtime": 96.396, | |
| "eval_samples_per_second": 19.129, | |
| "eval_steps_per_second": 2.396, | |
| "eval_wer": 0.4449934738019765, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5267, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "eval_loss": 0.3249629735946655, | |
| "eval_runtime": 105.1317, | |
| "eval_samples_per_second": 17.54, | |
| "eval_steps_per_second": 2.197, | |
| "eval_wer": 0.43660264777176955, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 16.94, | |
| "learning_rate": 0.0001, | |
| "loss": 0.49, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 16.94, | |
| "eval_loss": 0.2999746799468994, | |
| "eval_runtime": 97.027, | |
| "eval_samples_per_second": 19.005, | |
| "eval_steps_per_second": 2.381, | |
| "eval_wer": 0.4337124743613649, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 17.74, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4945, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 17.74, | |
| "eval_loss": 0.30671271681785583, | |
| "eval_runtime": 96.8855, | |
| "eval_samples_per_second": 19.033, | |
| "eval_steps_per_second": 2.384, | |
| "eval_wer": 0.43100876375163155, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "learning_rate": 0.0001, | |
| "loss": 0.485, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "eval_loss": 0.3134038746356964, | |
| "eval_runtime": 98.1951, | |
| "eval_samples_per_second": 18.779, | |
| "eval_steps_per_second": 2.352, | |
| "eval_wer": 0.43706880477344767, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 0.0001, | |
| "loss": 0.486, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "eval_loss": 0.3182927072048187, | |
| "eval_runtime": 100.6349, | |
| "eval_samples_per_second": 18.324, | |
| "eval_steps_per_second": 2.295, | |
| "eval_wer": 0.43044937534961775, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 20.16, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4617, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 20.16, | |
| "eval_loss": 0.3004864752292633, | |
| "eval_runtime": 102.1006, | |
| "eval_samples_per_second": 18.061, | |
| "eval_steps_per_second": 2.262, | |
| "eval_wer": 0.4238299459257878, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 20.97, | |
| "learning_rate": 0.0001, | |
| "loss": 0.465, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 20.97, | |
| "eval_loss": 0.313532292842865, | |
| "eval_runtime": 101.2873, | |
| "eval_samples_per_second": 18.206, | |
| "eval_steps_per_second": 2.281, | |
| "eval_wer": 0.4311019951519672, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 21.77, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4414, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 21.77, | |
| "eval_loss": 0.3118414580821991, | |
| "eval_runtime": 97.4069, | |
| "eval_samples_per_second": 18.931, | |
| "eval_steps_per_second": 2.371, | |
| "eval_wer": 0.4256013425321648, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 22.58, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4424, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 22.58, | |
| "eval_loss": 0.3086116313934326, | |
| "eval_runtime": 97.4809, | |
| "eval_samples_per_second": 18.917, | |
| "eval_steps_per_second": 2.37, | |
| "eval_wer": 0.4160917396979303, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 23.39, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4352, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 23.39, | |
| "eval_loss": 0.3213385045528412, | |
| "eval_runtime": 96.5991, | |
| "eval_samples_per_second": 19.089, | |
| "eval_steps_per_second": 2.391, | |
| "eval_wer": 0.41879545030766363, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 24.19, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4324, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 24.19, | |
| "eval_loss": 0.3011772036552429, | |
| "eval_runtime": 97.1915, | |
| "eval_samples_per_second": 18.973, | |
| "eval_steps_per_second": 2.377, | |
| "eval_wer": 0.4236434831251165, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4246, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.30194172263145447, | |
| "eval_runtime": 100.0292, | |
| "eval_samples_per_second": 18.435, | |
| "eval_steps_per_second": 2.309, | |
| "eval_wer": 0.4165578966996084, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4132, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "eval_loss": 0.299437552690506, | |
| "eval_runtime": 97.7952, | |
| "eval_samples_per_second": 18.856, | |
| "eval_steps_per_second": 2.362, | |
| "eval_wer": 0.4160917396979303, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 26.61, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4074, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 26.61, | |
| "eval_loss": 0.31787610054016113, | |
| "eval_runtime": 97.3925, | |
| "eval_samples_per_second": 18.934, | |
| "eval_steps_per_second": 2.372, | |
| "eval_wer": 0.41795636770464295, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 27.42, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4031, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 27.42, | |
| "eval_loss": 0.296748548746109, | |
| "eval_runtime": 101.7762, | |
| "eval_samples_per_second": 18.118, | |
| "eval_steps_per_second": 2.27, | |
| "eval_wer": 0.40760768226738764, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 28.23, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4082, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 28.23, | |
| "eval_loss": 0.31329259276390076, | |
| "eval_runtime": 102.7944, | |
| "eval_samples_per_second": 17.939, | |
| "eval_steps_per_second": 2.247, | |
| "eval_wer": 0.41310833488719, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 29.03, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3912, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 29.03, | |
| "eval_loss": 0.3259478509426117, | |
| "eval_runtime": 98.5616, | |
| "eval_samples_per_second": 18.709, | |
| "eval_steps_per_second": 2.344, | |
| "eval_wer": 0.41133693828081297, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 29.84, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3922, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 29.84, | |
| "eval_loss": 0.32406386733055115, | |
| "eval_runtime": 96.8005, | |
| "eval_samples_per_second": 19.049, | |
| "eval_steps_per_second": 2.386, | |
| "eval_wer": 0.41226925228416933, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 30.65, | |
| "learning_rate": 9.795698924731184e-05, | |
| "loss": 0.3851, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 30.65, | |
| "eval_loss": 0.29852330684661865, | |
| "eval_runtime": 105.0804, | |
| "eval_samples_per_second": 17.548, | |
| "eval_steps_per_second": 2.198, | |
| "eval_wer": 0.41133693828081297, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 31.45, | |
| "learning_rate": 9.526881720430108e-05, | |
| "loss": 0.3768, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 31.45, | |
| "eval_loss": 0.3223101794719696, | |
| "eval_runtime": 97.6458, | |
| "eval_samples_per_second": 18.885, | |
| "eval_steps_per_second": 2.366, | |
| "eval_wer": 0.41040462427745666, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "learning_rate": 9.258064516129033e-05, | |
| "loss": 0.3728, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "eval_loss": 0.3212699890136719, | |
| "eval_runtime": 100.4336, | |
| "eval_samples_per_second": 18.36, | |
| "eval_steps_per_second": 2.3, | |
| "eval_wer": 0.40770091366772326, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 33.06, | |
| "learning_rate": 8.989247311827958e-05, | |
| "loss": 0.3646, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 33.06, | |
| "eval_loss": 0.31770506501197815, | |
| "eval_runtime": 97.2462, | |
| "eval_samples_per_second": 18.962, | |
| "eval_steps_per_second": 2.375, | |
| "eval_wer": 0.4116166324818199, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 33.87, | |
| "learning_rate": 8.720430107526883e-05, | |
| "loss": 0.3681, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 33.87, | |
| "eval_loss": 0.3010982871055603, | |
| "eval_runtime": 99.148, | |
| "eval_samples_per_second": 18.598, | |
| "eval_steps_per_second": 2.33, | |
| "eval_wer": 0.4105910870781279, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 34.68, | |
| "learning_rate": 8.451612903225808e-05, | |
| "loss": 0.3634, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 34.68, | |
| "eval_loss": 0.3075733780860901, | |
| "eval_runtime": 99.8799, | |
| "eval_samples_per_second": 18.462, | |
| "eval_steps_per_second": 2.313, | |
| "eval_wer": 0.4069550624650382, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 35.48, | |
| "learning_rate": 8.182795698924732e-05, | |
| "loss": 0.3471, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 35.48, | |
| "eval_loss": 0.31304171681404114, | |
| "eval_runtime": 99.6768, | |
| "eval_samples_per_second": 18.5, | |
| "eval_steps_per_second": 2.317, | |
| "eval_wer": 0.40266641804959913, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 36.29, | |
| "learning_rate": 7.913978494623657e-05, | |
| "loss": 0.3398, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 36.29, | |
| "eval_loss": 0.306076318025589, | |
| "eval_runtime": 103.5936, | |
| "eval_samples_per_second": 17.8, | |
| "eval_steps_per_second": 2.23, | |
| "eval_wer": 0.4037851948536267, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 37.1, | |
| "learning_rate": 7.645161290322582e-05, | |
| "loss": 0.3394, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 37.1, | |
| "eval_loss": 0.31049424409866333, | |
| "eval_runtime": 98.5424, | |
| "eval_samples_per_second": 18.713, | |
| "eval_steps_per_second": 2.344, | |
| "eval_wer": 0.4037851948536267, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 37.9, | |
| "learning_rate": 7.376344086021507e-05, | |
| "loss": 0.331, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 37.9, | |
| "eval_loss": 0.3191221058368683, | |
| "eval_runtime": 102.0886, | |
| "eval_samples_per_second": 18.063, | |
| "eval_steps_per_second": 2.263, | |
| "eval_wer": 0.3992168562371807, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 38.71, | |
| "learning_rate": 7.10752688172043e-05, | |
| "loss": 0.3221, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 38.71, | |
| "eval_loss": 0.3119024336338043, | |
| "eval_runtime": 99.689, | |
| "eval_samples_per_second": 18.498, | |
| "eval_steps_per_second": 2.317, | |
| "eval_wer": 0.39958978183852323, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 39.52, | |
| "learning_rate": 6.838709677419355e-05, | |
| "loss": 0.3207, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 39.52, | |
| "eval_loss": 0.3276441991329193, | |
| "eval_runtime": 96.7884, | |
| "eval_samples_per_second": 19.052, | |
| "eval_steps_per_second": 2.387, | |
| "eval_wer": 0.40313257505127725, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 40.32, | |
| "learning_rate": 6.56989247311828e-05, | |
| "loss": 0.3186, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 40.32, | |
| "eval_loss": 0.3335840106010437, | |
| "eval_runtime": 98.1942, | |
| "eval_samples_per_second": 18.779, | |
| "eval_steps_per_second": 2.352, | |
| "eval_wer": 0.4032258064516129, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 41.13, | |
| "learning_rate": 6.301075268817205e-05, | |
| "loss": 0.3152, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 41.13, | |
| "eval_loss": 0.3129470944404602, | |
| "eval_runtime": 99.396, | |
| "eval_samples_per_second": 18.552, | |
| "eval_steps_per_second": 2.324, | |
| "eval_wer": 0.3941823606190565, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 41.94, | |
| "learning_rate": 6.0322580645161295e-05, | |
| "loss": 0.3093, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 41.94, | |
| "eval_loss": 0.31502029299736023, | |
| "eval_runtime": 97.1686, | |
| "eval_samples_per_second": 18.977, | |
| "eval_steps_per_second": 2.377, | |
| "eval_wer": 0.3930635838150289, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 42.74, | |
| "learning_rate": 5.763440860215054e-05, | |
| "loss": 0.2949, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 42.74, | |
| "eval_loss": 0.3235335052013397, | |
| "eval_runtime": 102.899, | |
| "eval_samples_per_second": 17.92, | |
| "eval_steps_per_second": 2.245, | |
| "eval_wer": 0.3953011374230841, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 43.55, | |
| "learning_rate": 5.494623655913979e-05, | |
| "loss": 0.3013, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 43.55, | |
| "eval_loss": 0.3291964530944824, | |
| "eval_runtime": 97.4704, | |
| "eval_samples_per_second": 18.919, | |
| "eval_steps_per_second": 2.37, | |
| "eval_wer": 0.3934365094163714, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 44.35, | |
| "learning_rate": 5.225806451612903e-05, | |
| "loss": 0.3092, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 44.35, | |
| "eval_loss": 0.3066520392894745, | |
| "eval_runtime": 96.9166, | |
| "eval_samples_per_second": 19.027, | |
| "eval_steps_per_second": 2.383, | |
| "eval_wer": 0.38616446019019207, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 45.16, | |
| "learning_rate": 4.956989247311828e-05, | |
| "loss": 0.2938, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 45.16, | |
| "eval_loss": 0.32365530729293823, | |
| "eval_runtime": 98.8953, | |
| "eval_samples_per_second": 18.646, | |
| "eval_steps_per_second": 2.336, | |
| "eval_wer": 0.38756293119522656, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 45.97, | |
| "learning_rate": 4.688172043010753e-05, | |
| "loss": 0.2985, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 45.97, | |
| "eval_loss": 0.3358381390571594, | |
| "eval_runtime": 96.7245, | |
| "eval_samples_per_second": 19.064, | |
| "eval_steps_per_second": 2.388, | |
| "eval_wer": 0.3876561625955622, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 46.77, | |
| "learning_rate": 4.4193548387096775e-05, | |
| "loss": 0.2943, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 46.77, | |
| "eval_loss": 0.3283197283744812, | |
| "eval_runtime": 98.5499, | |
| "eval_samples_per_second": 18.711, | |
| "eval_steps_per_second": 2.344, | |
| "eval_wer": 0.3928771210143576, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 47.58, | |
| "learning_rate": 4.1505376344086025e-05, | |
| "loss": 0.2754, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 47.58, | |
| "eval_loss": 0.33124613761901855, | |
| "eval_runtime": 97.2557, | |
| "eval_samples_per_second": 18.96, | |
| "eval_steps_per_second": 2.375, | |
| "eval_wer": 0.38952079060227485, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 48.39, | |
| "learning_rate": 3.881720430107527e-05, | |
| "loss": 0.2906, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 48.39, | |
| "eval_loss": 0.33012691140174866, | |
| "eval_runtime": 99.7418, | |
| "eval_samples_per_second": 18.488, | |
| "eval_steps_per_second": 2.316, | |
| "eval_wer": 0.39045310460563115, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 49.19, | |
| "learning_rate": 3.612903225806452e-05, | |
| "loss": 0.2751, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 49.19, | |
| "eval_loss": 0.3261188566684723, | |
| "eval_runtime": 98.7276, | |
| "eval_samples_per_second": 18.678, | |
| "eval_steps_per_second": 2.34, | |
| "eval_wer": 0.3873764683945553, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 3.344086021505377e-05, | |
| "loss": 0.2825, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 0.3322080671787262, | |
| "eval_runtime": 97.0188, | |
| "eval_samples_per_second": 19.007, | |
| "eval_steps_per_second": 2.381, | |
| "eval_wer": 0.38700354279321275, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 50.81, | |
| "learning_rate": 3.075268817204301e-05, | |
| "loss": 0.2727, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 50.81, | |
| "eval_loss": 0.3448057770729065, | |
| "eval_runtime": 98.9754, | |
| "eval_samples_per_second": 18.631, | |
| "eval_steps_per_second": 2.334, | |
| "eval_wer": 0.3928771210143576, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 51.61, | |
| "learning_rate": 2.806451612903226e-05, | |
| "loss": 0.2896, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 51.61, | |
| "eval_loss": 0.33276790380477905, | |
| "eval_runtime": 96.9108, | |
| "eval_samples_per_second": 19.028, | |
| "eval_steps_per_second": 2.384, | |
| "eval_wer": 0.38858847659891854, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 52.42, | |
| "learning_rate": 2.537634408602151e-05, | |
| "loss": 0.2753, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 52.42, | |
| "eval_loss": 0.32856670022010803, | |
| "eval_runtime": 105.68, | |
| "eval_samples_per_second": 17.449, | |
| "eval_steps_per_second": 2.186, | |
| "eval_wer": 0.3876561625955622, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 53.23, | |
| "learning_rate": 2.268817204301075e-05, | |
| "loss": 0.271, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 53.23, | |
| "eval_loss": 0.3287724256515503, | |
| "eval_runtime": 100.7427, | |
| "eval_samples_per_second": 18.304, | |
| "eval_steps_per_second": 2.293, | |
| "eval_wer": 0.3876561625955622, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 54.03, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2664, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 54.03, | |
| "eval_loss": 0.3290824890136719, | |
| "eval_runtime": 98.5577, | |
| "eval_samples_per_second": 18.71, | |
| "eval_steps_per_second": 2.344, | |
| "eval_wer": 0.3893343278016036, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 54.84, | |
| "learning_rate": 1.7311827956989248e-05, | |
| "loss": 0.2491, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 54.84, | |
| "eval_loss": 0.33568552136421204, | |
| "eval_runtime": 103.2565, | |
| "eval_samples_per_second": 17.858, | |
| "eval_steps_per_second": 2.237, | |
| "eval_wer": 0.388215550997576, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 55.65, | |
| "learning_rate": 1.4623655913978495e-05, | |
| "loss": 0.2569, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 55.65, | |
| "eval_loss": 0.3253258466720581, | |
| "eval_runtime": 104.0214, | |
| "eval_samples_per_second": 17.727, | |
| "eval_steps_per_second": 2.221, | |
| "eval_wer": 0.3876561625955622, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 56.45, | |
| "learning_rate": 1.1935483870967743e-05, | |
| "loss": 0.2568, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 56.45, | |
| "eval_loss": 0.3199854791164398, | |
| "eval_runtime": 103.944, | |
| "eval_samples_per_second": 17.74, | |
| "eval_steps_per_second": 2.222, | |
| "eval_wer": 0.3868170799925415, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 57.26, | |
| "learning_rate": 9.24731182795699e-06, | |
| "loss": 0.2668, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 57.26, | |
| "eval_loss": 0.32368558645248413, | |
| "eval_runtime": 103.5633, | |
| "eval_samples_per_second": 17.806, | |
| "eval_steps_per_second": 2.231, | |
| "eval_wer": 0.3862576915905277, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 58.06, | |
| "learning_rate": 6.586021505376344e-06, | |
| "loss": 0.253, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 58.06, | |
| "eval_loss": 0.3248105049133301, | |
| "eval_runtime": 103.7822, | |
| "eval_samples_per_second": 17.768, | |
| "eval_steps_per_second": 2.226, | |
| "eval_wer": 0.38504568338616446, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 58.87, | |
| "learning_rate": 3.8978494623655915e-06, | |
| "loss": 0.2621, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 58.87, | |
| "eval_loss": 0.3238443434238434, | |
| "eval_runtime": 97.9464, | |
| "eval_samples_per_second": 18.827, | |
| "eval_steps_per_second": 2.358, | |
| "eval_wer": 0.38448629498415066, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 59.68, | |
| "learning_rate": 1.2096774193548388e-06, | |
| "loss": 0.2571, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 59.68, | |
| "eval_loss": 0.3273279666900635, | |
| "eval_runtime": 101.5169, | |
| "eval_samples_per_second": 18.164, | |
| "eval_steps_per_second": 2.275, | |
| "eval_wer": 0.3852321461868357, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "step": 7440, | |
| "total_flos": 2.7388066347434852e+19, | |
| "train_loss": 0.8272683016715511, | |
| "train_runtime": 19767.0967, | |
| "train_samples_per_second": 12.035, | |
| "train_steps_per_second": 0.376 | |
| } | |
| ], | |
| "max_steps": 7440, | |
| "num_train_epochs": 60, | |
| "total_flos": 2.7388066347434852e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |