| { | |
| "best_metric": 0.7032522559165955, | |
| "best_model_checkpoint": "./Hubert-noisy_common_voice_debug/checkpoint-3700", | |
| "epoch": 30.0, | |
| "eval_steps": 100, | |
| "global_step": 11280, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.26595744680851063, | |
| "eval_cer": 1.0538678168650442, | |
| "eval_loss": 12.082180976867676, | |
| "eval_runtime": 151.0989, | |
| "eval_samples_per_second": 32.833, | |
| "eval_steps_per_second": 4.11, | |
| "eval_wer": 1.1570247933884297, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5319148936170213, | |
| "eval_cer": 0.9816708830000044, | |
| "eval_loss": 5.878890037536621, | |
| "eval_runtime": 151.597, | |
| "eval_samples_per_second": 32.725, | |
| "eval_steps_per_second": 4.096, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7978723404255319, | |
| "eval_cer": 0.9816665437204115, | |
| "eval_loss": 5.362728595733643, | |
| "eval_runtime": 149.1154, | |
| "eval_samples_per_second": 33.27, | |
| "eval_steps_per_second": 4.165, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0638297872340425, | |
| "eval_cer": 0.9816752222795971, | |
| "eval_loss": 4.9316487312316895, | |
| "eval_runtime": 149.2108, | |
| "eval_samples_per_second": 33.248, | |
| "eval_steps_per_second": 4.162, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.3297872340425532, | |
| "grad_norm": 18.269367218017578, | |
| "learning_rate": 1.1927999999999998e-05, | |
| "loss": 6.372, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.3297872340425532, | |
| "eval_cer": 0.9816752222795971, | |
| "eval_loss": 4.455592632293701, | |
| "eval_runtime": 149.4229, | |
| "eval_samples_per_second": 33.201, | |
| "eval_steps_per_second": 4.156, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5957446808510638, | |
| "eval_cer": 0.9816752222795971, | |
| "eval_loss": 3.989032506942749, | |
| "eval_runtime": 148.7951, | |
| "eval_samples_per_second": 33.341, | |
| "eval_steps_per_second": 4.174, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8617021276595744, | |
| "eval_cer": 0.9816622044408188, | |
| "eval_loss": 3.57344126701355, | |
| "eval_runtime": 148.5583, | |
| "eval_samples_per_second": 33.394, | |
| "eval_steps_per_second": 4.18, | |
| "eval_wer": 1.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "eval_cer": 0.9816708830000044, | |
| "eval_loss": 3.2931714057922363, | |
| "eval_runtime": 148.3742, | |
| "eval_samples_per_second": 33.436, | |
| "eval_steps_per_second": 4.185, | |
| "eval_wer": 1.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.393617021276596, | |
| "eval_cer": 0.9816839008387828, | |
| "eval_loss": 3.1536319255828857, | |
| "eval_runtime": 148.8499, | |
| "eval_samples_per_second": 33.329, | |
| "eval_steps_per_second": 4.172, | |
| "eval_wer": 1.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "grad_norm": 1.548904299736023, | |
| "learning_rate": 2.3928e-05, | |
| "loss": 3.4101, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "eval_cer": 0.9816622044408188, | |
| "eval_loss": 3.0483851432800293, | |
| "eval_runtime": 148.9657, | |
| "eval_samples_per_second": 33.303, | |
| "eval_steps_per_second": 4.169, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.925531914893617, | |
| "eval_cer": 0.9807552950059231, | |
| "eval_loss": 2.8469536304473877, | |
| "eval_runtime": 148.996, | |
| "eval_samples_per_second": 33.296, | |
| "eval_steps_per_second": 4.168, | |
| "eval_wer": 1.0, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.1914893617021276, | |
| "eval_cer": 0.8702078080996993, | |
| "eval_loss": 2.521139144897461, | |
| "eval_runtime": 149.4933, | |
| "eval_samples_per_second": 33.185, | |
| "eval_steps_per_second": 4.154, | |
| "eval_wer": 1.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.4574468085106385, | |
| "eval_cer": 0.5518348643758163, | |
| "eval_loss": 2.035446882247925, | |
| "eval_runtime": 150.325, | |
| "eval_samples_per_second": 33.002, | |
| "eval_steps_per_second": 4.131, | |
| "eval_wer": 1.0, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.723404255319149, | |
| "eval_cer": 0.44136548450226293, | |
| "eval_loss": 1.678019404411316, | |
| "eval_runtime": 150.3487, | |
| "eval_samples_per_second": 32.997, | |
| "eval_steps_per_second": 4.13, | |
| "eval_wer": 1.0, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "grad_norm": 4.6614580154418945, | |
| "learning_rate": 3.5928e-05, | |
| "loss": 2.3222, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "eval_cer": 0.43124194521225584, | |
| "eval_loss": 1.5038928985595703, | |
| "eval_runtime": 150.8122, | |
| "eval_samples_per_second": 32.895, | |
| "eval_steps_per_second": 4.118, | |
| "eval_wer": 1.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.25531914893617, | |
| "eval_cer": 0.396458279996355, | |
| "eval_loss": 1.3418982028961182, | |
| "eval_runtime": 150.4572, | |
| "eval_samples_per_second": 32.973, | |
| "eval_steps_per_second": 4.127, | |
| "eval_wer": 1.0, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.5212765957446805, | |
| "eval_cer": 0.3685610514942309, | |
| "eval_loss": 1.2053722143173218, | |
| "eval_runtime": 150.9364, | |
| "eval_samples_per_second": 32.868, | |
| "eval_steps_per_second": 4.114, | |
| "eval_wer": 1.0, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 4.787234042553192, | |
| "eval_cer": 0.3321067636351013, | |
| "eval_loss": 1.0588372945785522, | |
| "eval_runtime": 150.9127, | |
| "eval_samples_per_second": 32.873, | |
| "eval_steps_per_second": 4.115, | |
| "eval_wer": 1.0, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.053191489361702, | |
| "eval_cer": 0.31576069740901613, | |
| "eval_loss": 0.9545913338661194, | |
| "eval_runtime": 150.8858, | |
| "eval_samples_per_second": 32.879, | |
| "eval_steps_per_second": 4.116, | |
| "eval_wer": 1.0, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 5.319148936170213, | |
| "grad_norm": 4.519205570220947, | |
| "learning_rate": 4.7928e-05, | |
| "loss": 1.2343, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.319148936170213, | |
| "eval_cer": 0.3105969546935818, | |
| "eval_loss": 0.9041823148727417, | |
| "eval_runtime": 150.7252, | |
| "eval_samples_per_second": 32.914, | |
| "eval_steps_per_second": 4.12, | |
| "eval_wer": 1.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.585106382978723, | |
| "eval_cer": 0.30879615366256896, | |
| "eval_loss": 0.8747320771217346, | |
| "eval_runtime": 151.661, | |
| "eval_samples_per_second": 32.711, | |
| "eval_steps_per_second": 4.095, | |
| "eval_wer": 1.0, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 5.851063829787234, | |
| "eval_cer": 0.2971885807518236, | |
| "eval_loss": 0.8223557472229004, | |
| "eval_runtime": 155.0175, | |
| "eval_samples_per_second": 32.003, | |
| "eval_steps_per_second": 4.006, | |
| "eval_wer": 1.0, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 6.117021276595745, | |
| "eval_cer": 0.2995751845278647, | |
| "eval_loss": 0.8100646138191223, | |
| "eval_runtime": 151.2405, | |
| "eval_samples_per_second": 32.802, | |
| "eval_steps_per_second": 4.106, | |
| "eval_wer": 1.0, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 6.382978723404255, | |
| "eval_cer": 0.2969933131701475, | |
| "eval_loss": 0.7891868948936462, | |
| "eval_runtime": 151.3984, | |
| "eval_samples_per_second": 32.768, | |
| "eval_steps_per_second": 4.102, | |
| "eval_wer": 1.0, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.648936170212766, | |
| "grad_norm": 7.927753925323486, | |
| "learning_rate": 5.9903999999999994e-05, | |
| "loss": 0.8716, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.648936170212766, | |
| "eval_cer": 0.29149544592606735, | |
| "eval_loss": 0.7661293148994446, | |
| "eval_runtime": 152.5094, | |
| "eval_samples_per_second": 32.529, | |
| "eval_steps_per_second": 4.072, | |
| "eval_wer": 1.0, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.914893617021277, | |
| "eval_cer": 0.28864019995400364, | |
| "eval_loss": 0.7654036283493042, | |
| "eval_runtime": 151.4659, | |
| "eval_samples_per_second": 32.753, | |
| "eval_steps_per_second": 4.1, | |
| "eval_wer": 1.0, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 7.180851063829787, | |
| "eval_cer": 0.28976841264813213, | |
| "eval_loss": 0.767704963684082, | |
| "eval_runtime": 151.3213, | |
| "eval_samples_per_second": 32.785, | |
| "eval_steps_per_second": 4.104, | |
| "eval_wer": 1.0, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 7.446808510638298, | |
| "eval_cer": 0.28610606067180727, | |
| "eval_loss": 0.752834141254425, | |
| "eval_runtime": 151.5246, | |
| "eval_samples_per_second": 32.741, | |
| "eval_steps_per_second": 4.098, | |
| "eval_wer": 1.0, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.712765957446808, | |
| "eval_cer": 0.28800232585386176, | |
| "eval_loss": 0.7432949542999268, | |
| "eval_runtime": 155.4404, | |
| "eval_samples_per_second": 31.916, | |
| "eval_steps_per_second": 3.995, | |
| "eval_wer": 1.0, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 7.9787234042553195, | |
| "grad_norm": 6.501748085021973, | |
| "learning_rate": 7.1904e-05, | |
| "loss": 0.7324, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.9787234042553195, | |
| "eval_cer": 0.2876595227660304, | |
| "eval_loss": 0.7498476505279541, | |
| "eval_runtime": 151.1874, | |
| "eval_samples_per_second": 32.814, | |
| "eval_steps_per_second": 4.107, | |
| "eval_wer": 1.0, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.24468085106383, | |
| "eval_cer": 0.2827474582669785, | |
| "eval_loss": 0.7267456650733948, | |
| "eval_runtime": 151.7111, | |
| "eval_samples_per_second": 32.7, | |
| "eval_steps_per_second": 4.093, | |
| "eval_wer": 1.0, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 8.51063829787234, | |
| "eval_cer": 0.28125474608705464, | |
| "eval_loss": 0.7319472432136536, | |
| "eval_runtime": 151.6994, | |
| "eval_samples_per_second": 32.703, | |
| "eval_steps_per_second": 4.094, | |
| "eval_wer": 1.0, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 8.77659574468085, | |
| "eval_cer": 0.28822796839268744, | |
| "eval_loss": 0.7477809190750122, | |
| "eval_runtime": 151.7004, | |
| "eval_samples_per_second": 32.703, | |
| "eval_steps_per_second": 4.094, | |
| "eval_wer": 1.0, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 9.042553191489361, | |
| "eval_cer": 0.2814543529483235, | |
| "eval_loss": 0.7336912751197815, | |
| "eval_runtime": 151.8383, | |
| "eval_samples_per_second": 32.673, | |
| "eval_steps_per_second": 4.09, | |
| "eval_wer": 1.0, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 9.308510638297872, | |
| "grad_norm": 6.825859546661377, | |
| "learning_rate": 8.390399999999999e-05, | |
| "loss": 0.6486, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.308510638297872, | |
| "eval_cer": 0.28514274060220524, | |
| "eval_loss": 0.734130322933197, | |
| "eval_runtime": 151.4163, | |
| "eval_samples_per_second": 32.764, | |
| "eval_steps_per_second": 4.101, | |
| "eval_wer": 1.0, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.574468085106384, | |
| "eval_cer": 0.2803218009746022, | |
| "eval_loss": 0.74192875623703, | |
| "eval_runtime": 151.3965, | |
| "eval_samples_per_second": 32.768, | |
| "eval_steps_per_second": 4.102, | |
| "eval_wer": 1.0, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 9.840425531914894, | |
| "eval_cer": 0.277323358775976, | |
| "eval_loss": 0.7032522559165955, | |
| "eval_runtime": 151.1826, | |
| "eval_samples_per_second": 32.815, | |
| "eval_steps_per_second": 4.108, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 10.106382978723405, | |
| "eval_cer": 0.2829253687302834, | |
| "eval_loss": 0.7326785922050476, | |
| "eval_runtime": 151.1178, | |
| "eval_samples_per_second": 32.829, | |
| "eval_steps_per_second": 4.109, | |
| "eval_wer": 1.0, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 10.372340425531915, | |
| "eval_cer": 0.2855332757655574, | |
| "eval_loss": 0.7554399967193604, | |
| "eval_runtime": 151.3763, | |
| "eval_samples_per_second": 32.773, | |
| "eval_steps_per_second": 4.102, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 10.638297872340425, | |
| "grad_norm": 5.65977144241333, | |
| "learning_rate": 9.5904e-05, | |
| "loss": 0.6034, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.638297872340425, | |
| "eval_cer": 0.2841360277366752, | |
| "eval_loss": 0.7361173033714294, | |
| "eval_runtime": 151.2103, | |
| "eval_samples_per_second": 32.809, | |
| "eval_steps_per_second": 4.107, | |
| "eval_wer": 1.0, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.904255319148936, | |
| "eval_cer": 0.2833202431732284, | |
| "eval_loss": 0.745939314365387, | |
| "eval_runtime": 151.41, | |
| "eval_samples_per_second": 32.765, | |
| "eval_steps_per_second": 4.101, | |
| "eval_wer": 1.0, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 11.170212765957446, | |
| "eval_cer": 0.2801352119521117, | |
| "eval_loss": 0.7383919358253479, | |
| "eval_runtime": 151.4933, | |
| "eval_samples_per_second": 32.747, | |
| "eval_steps_per_second": 4.099, | |
| "eval_wer": 1.0, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 11.436170212765958, | |
| "eval_cer": 0.27755767987398733, | |
| "eval_loss": 0.7336695194244385, | |
| "eval_runtime": 151.6319, | |
| "eval_samples_per_second": 32.717, | |
| "eval_steps_per_second": 4.095, | |
| "eval_wer": 1.0, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 11.702127659574469, | |
| "eval_cer": 0.2819099773055677, | |
| "eval_loss": 0.7572408318519592, | |
| "eval_runtime": 151.2617, | |
| "eval_samples_per_second": 32.797, | |
| "eval_steps_per_second": 4.105, | |
| "eval_wer": 1.0, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 11.96808510638298, | |
| "grad_norm": 5.250299453735352, | |
| "learning_rate": 0.00010790399999999999, | |
| "loss": 0.5687, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 11.96808510638298, | |
| "eval_cer": 0.2824393694158896, | |
| "eval_loss": 0.7522115707397461, | |
| "eval_runtime": 151.5116, | |
| "eval_samples_per_second": 32.743, | |
| "eval_steps_per_second": 4.099, | |
| "eval_wer": 1.0, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.23404255319149, | |
| "eval_cer": 0.27894624934368395, | |
| "eval_loss": 0.7491241097450256, | |
| "eval_runtime": 151.5087, | |
| "eval_samples_per_second": 32.744, | |
| "eval_steps_per_second": 4.099, | |
| "eval_wer": 1.0, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "eval_cer": 0.2831987433446299, | |
| "eval_loss": 0.7484813928604126, | |
| "eval_runtime": 151.4358, | |
| "eval_samples_per_second": 32.76, | |
| "eval_steps_per_second": 4.101, | |
| "eval_wer": 1.0, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 12.76595744680851, | |
| "eval_cer": 0.2849040802246011, | |
| "eval_loss": 0.7622763514518738, | |
| "eval_runtime": 151.3314, | |
| "eval_samples_per_second": 32.782, | |
| "eval_steps_per_second": 4.104, | |
| "eval_wer": 1.0, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 13.03191489361702, | |
| "eval_cer": 0.285936828767688, | |
| "eval_loss": 0.7829335331916809, | |
| "eval_runtime": 151.6153, | |
| "eval_samples_per_second": 32.721, | |
| "eval_steps_per_second": 4.096, | |
| "eval_wer": 1.0, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 13.297872340425531, | |
| "grad_norm": 4.461331844329834, | |
| "learning_rate": 0.00011990399999999998, | |
| "loss": 0.5255, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.297872340425531, | |
| "eval_cer": 0.28200110217701657, | |
| "eval_loss": 0.7818682789802551, | |
| "eval_runtime": 151.2326, | |
| "eval_samples_per_second": 32.804, | |
| "eval_steps_per_second": 4.106, | |
| "eval_wer": 1.0, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.563829787234042, | |
| "eval_cer": 0.2824393694158896, | |
| "eval_loss": 0.7782630324363708, | |
| "eval_runtime": 151.6654, | |
| "eval_samples_per_second": 32.71, | |
| "eval_steps_per_second": 4.095, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 13.829787234042554, | |
| "eval_cer": 0.2840275457468551, | |
| "eval_loss": 0.7653459906578064, | |
| "eval_runtime": 151.0873, | |
| "eval_samples_per_second": 32.835, | |
| "eval_steps_per_second": 4.11, | |
| "eval_wer": 1.0, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 14.095744680851064, | |
| "eval_cer": 0.28220938759747105, | |
| "eval_loss": 0.7816305756568909, | |
| "eval_runtime": 151.1619, | |
| "eval_samples_per_second": 32.819, | |
| "eval_steps_per_second": 4.108, | |
| "eval_wer": 1.0, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 14.361702127659575, | |
| "eval_cer": 0.2824133337383328, | |
| "eval_loss": 0.7607858180999756, | |
| "eval_runtime": 150.8484, | |
| "eval_samples_per_second": 32.887, | |
| "eval_steps_per_second": 4.117, | |
| "eval_wer": 1.0, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 14.627659574468085, | |
| "grad_norm": 4.656615257263184, | |
| "learning_rate": 0.000131904, | |
| "loss": 0.5016, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.627659574468085, | |
| "eval_cer": 0.28411433133871117, | |
| "eval_loss": 0.7712321281433105, | |
| "eval_runtime": 151.045, | |
| "eval_samples_per_second": 32.845, | |
| "eval_steps_per_second": 4.111, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.893617021276595, | |
| "eval_cer": 0.286396792404525, | |
| "eval_loss": 0.7712020874023438, | |
| "eval_runtime": 151.0967, | |
| "eval_samples_per_second": 32.833, | |
| "eval_steps_per_second": 4.11, | |
| "eval_wer": 1.0, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 15.159574468085106, | |
| "eval_cer": 0.285060294289942, | |
| "eval_loss": 0.8153163194656372, | |
| "eval_runtime": 151.4841, | |
| "eval_samples_per_second": 32.749, | |
| "eval_steps_per_second": 4.099, | |
| "eval_wer": 0.999596855472687, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 15.425531914893616, | |
| "eval_cer": 0.2852338654736541, | |
| "eval_loss": 0.8161126375198364, | |
| "eval_runtime": 150.9671, | |
| "eval_samples_per_second": 32.861, | |
| "eval_steps_per_second": 4.113, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 15.691489361702128, | |
| "eval_cer": 0.28826702190902265, | |
| "eval_loss": 0.7910758852958679, | |
| "eval_runtime": 151.0378, | |
| "eval_samples_per_second": 32.846, | |
| "eval_steps_per_second": 4.112, | |
| "eval_wer": 1.0, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 15.957446808510639, | |
| "grad_norm": 5.0891499519348145, | |
| "learning_rate": 0.00014390399999999998, | |
| "loss": 0.4821, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.957446808510639, | |
| "eval_cer": 0.28232654814647673, | |
| "eval_loss": 0.7926108241081238, | |
| "eval_runtime": 151.2593, | |
| "eval_samples_per_second": 32.798, | |
| "eval_steps_per_second": 4.106, | |
| "eval_wer": 1.0, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 16.22340425531915, | |
| "eval_cer": 0.28674393477194915, | |
| "eval_loss": 0.814721405506134, | |
| "eval_runtime": 150.8029, | |
| "eval_samples_per_second": 32.897, | |
| "eval_steps_per_second": 4.118, | |
| "eval_wer": 1.0, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 16.48936170212766, | |
| "eval_cer": 0.28262161915878725, | |
| "eval_loss": 0.7699744701385498, | |
| "eval_runtime": 150.7429, | |
| "eval_samples_per_second": 32.91, | |
| "eval_steps_per_second": 4.12, | |
| "eval_wer": 1.0, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 16.75531914893617, | |
| "eval_cer": 0.2909530359769671, | |
| "eval_loss": 0.8119201064109802, | |
| "eval_runtime": 150.6966, | |
| "eval_samples_per_second": 32.92, | |
| "eval_steps_per_second": 4.121, | |
| "eval_wer": 1.0, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 17.02127659574468, | |
| "eval_cer": 0.2845873128143266, | |
| "eval_loss": 0.835509181022644, | |
| "eval_runtime": 151.1321, | |
| "eval_samples_per_second": 32.826, | |
| "eval_steps_per_second": 4.109, | |
| "eval_wer": 1.0, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 17.28723404255319, | |
| "grad_norm": 4.005585670471191, | |
| "learning_rate": 0.00015587999999999998, | |
| "loss": 0.4503, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.28723404255319, | |
| "eval_cer": 0.28592381092890956, | |
| "eval_loss": 0.793612539768219, | |
| "eval_runtime": 150.9291, | |
| "eval_samples_per_second": 32.87, | |
| "eval_steps_per_second": 4.115, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.5531914893617, | |
| "eval_cer": 0.29515779790239227, | |
| "eval_loss": 0.7976419925689697, | |
| "eval_runtime": 169.2668, | |
| "eval_samples_per_second": 29.309, | |
| "eval_steps_per_second": 3.669, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 17.819148936170212, | |
| "eval_cer": 0.2901546085318915, | |
| "eval_loss": 0.8274039626121521, | |
| "eval_runtime": 158.3816, | |
| "eval_samples_per_second": 31.323, | |
| "eval_steps_per_second": 3.921, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 18.085106382978722, | |
| "eval_cer": 0.2885490750825548, | |
| "eval_loss": 0.903409481048584, | |
| "eval_runtime": 151.19, | |
| "eval_samples_per_second": 32.813, | |
| "eval_steps_per_second": 4.107, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 18.351063829787233, | |
| "eval_cer": 0.28819325415594504, | |
| "eval_loss": 0.8065901398658752, | |
| "eval_runtime": 152.3883, | |
| "eval_samples_per_second": 32.555, | |
| "eval_steps_per_second": 4.075, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 18.617021276595743, | |
| "grad_norm": 4.851210117340088, | |
| "learning_rate": 0.000167856, | |
| "loss": 0.4435, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 18.617021276595743, | |
| "eval_cer": 0.29208124867109564, | |
| "eval_loss": 0.849502682685852, | |
| "eval_runtime": 171.4215, | |
| "eval_samples_per_second": 28.94, | |
| "eval_steps_per_second": 3.623, | |
| "eval_wer": 1.0, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 18.882978723404257, | |
| "eval_cer": 0.2896121985827913, | |
| "eval_loss": 0.8448098301887512, | |
| "eval_runtime": 151.0183, | |
| "eval_samples_per_second": 32.85, | |
| "eval_steps_per_second": 4.112, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 19.148936170212767, | |
| "eval_cer": 0.29039760818908844, | |
| "eval_loss": 0.877412736415863, | |
| "eval_runtime": 155.1268, | |
| "eval_samples_per_second": 31.98, | |
| "eval_steps_per_second": 4.003, | |
| "eval_wer": 1.0, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 19.414893617021278, | |
| "eval_cer": 0.29732309841920046, | |
| "eval_loss": 0.8293471336364746, | |
| "eval_runtime": 151.2836, | |
| "eval_samples_per_second": 32.793, | |
| "eval_steps_per_second": 4.105, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 19.680851063829788, | |
| "eval_cer": 0.2924804623936334, | |
| "eval_loss": 0.8037762641906738, | |
| "eval_runtime": 151.1502, | |
| "eval_samples_per_second": 32.822, | |
| "eval_steps_per_second": 4.108, | |
| "eval_wer": 1.0, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 19.9468085106383, | |
| "grad_norm": 3.5915985107421875, | |
| "learning_rate": 0.000179856, | |
| "loss": 0.4457, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 19.9468085106383, | |
| "eval_cer": 0.29077512551366225, | |
| "eval_loss": 0.806207537651062, | |
| "eval_runtime": 154.4071, | |
| "eval_samples_per_second": 32.129, | |
| "eval_steps_per_second": 4.022, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 20.21276595744681, | |
| "eval_cer": 0.29176448126082105, | |
| "eval_loss": 0.8740444779396057, | |
| "eval_runtime": 154.5779, | |
| "eval_samples_per_second": 32.094, | |
| "eval_steps_per_second": 4.017, | |
| "eval_wer": 1.0, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 20.47872340425532, | |
| "eval_cer": 0.2977179728621454, | |
| "eval_loss": 0.848852276802063, | |
| "eval_runtime": 173.0214, | |
| "eval_samples_per_second": 28.673, | |
| "eval_steps_per_second": 3.589, | |
| "eval_wer": 1.0, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 20.74468085106383, | |
| "eval_cer": 0.29728404490286525, | |
| "eval_loss": 0.8606219291687012, | |
| "eval_runtime": 151.0098, | |
| "eval_samples_per_second": 32.852, | |
| "eval_steps_per_second": 4.112, | |
| "eval_wer": 1.0, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 21.01063829787234, | |
| "eval_cer": 0.29262799789978866, | |
| "eval_loss": 0.8141107559204102, | |
| "eval_runtime": 155.5633, | |
| "eval_samples_per_second": 31.891, | |
| "eval_steps_per_second": 3.992, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 21.27659574468085, | |
| "grad_norm": 5.225797176361084, | |
| "learning_rate": 0.000191856, | |
| "loss": 0.4252, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 21.27659574468085, | |
| "eval_cer": 0.29841225759699375, | |
| "eval_loss": 0.8832473158836365, | |
| "eval_runtime": 161.5174, | |
| "eval_samples_per_second": 30.715, | |
| "eval_steps_per_second": 3.845, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 21.54255319148936, | |
| "eval_cer": 0.29451124524306477, | |
| "eval_loss": 0.8589781522750854, | |
| "eval_runtime": 150.7991, | |
| "eval_samples_per_second": 32.898, | |
| "eval_steps_per_second": 4.118, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 21.80851063829787, | |
| "eval_cer": 0.2939644960143717, | |
| "eval_loss": 0.8304488062858582, | |
| "eval_runtime": 156.9786, | |
| "eval_samples_per_second": 31.603, | |
| "eval_steps_per_second": 3.956, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 22.074468085106382, | |
| "eval_cer": 0.29739686617227806, | |
| "eval_loss": 0.873356819152832, | |
| "eval_runtime": 151.6885, | |
| "eval_samples_per_second": 32.705, | |
| "eval_steps_per_second": 4.094, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 22.340425531914892, | |
| "eval_cer": 0.292992497385584, | |
| "eval_loss": 0.8416581153869629, | |
| "eval_runtime": 171.0572, | |
| "eval_samples_per_second": 29.002, | |
| "eval_steps_per_second": 3.63, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 22.606382978723403, | |
| "grad_norm": 6.305420875549316, | |
| "learning_rate": 0.00020385599999999998, | |
| "loss": 0.418, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 22.606382978723403, | |
| "eval_cer": 0.2993235063114822, | |
| "eval_loss": 0.9387206435203552, | |
| "eval_runtime": 154.7978, | |
| "eval_samples_per_second": 32.048, | |
| "eval_steps_per_second": 4.012, | |
| "eval_wer": 1.0, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 22.872340425531917, | |
| "eval_cer": 0.2996055594850143, | |
| "eval_loss": 0.8810231685638428, | |
| "eval_runtime": 154.5989, | |
| "eval_samples_per_second": 32.089, | |
| "eval_steps_per_second": 4.017, | |
| "eval_wer": 1.0, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 23.138297872340427, | |
| "eval_cer": 0.3074119234724651, | |
| "eval_loss": 0.9089604020118713, | |
| "eval_runtime": 151.1053, | |
| "eval_samples_per_second": 32.831, | |
| "eval_steps_per_second": 4.11, | |
| "eval_wer": 1.0, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 23.404255319148938, | |
| "eval_cer": 0.31074015092014423, | |
| "eval_loss": 0.8992817997932434, | |
| "eval_runtime": 154.115, | |
| "eval_samples_per_second": 32.19, | |
| "eval_steps_per_second": 4.029, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 23.670212765957448, | |
| "eval_cer": 0.3032852685797104, | |
| "eval_loss": 0.8724159598350525, | |
| "eval_runtime": 152.578, | |
| "eval_samples_per_second": 32.515, | |
| "eval_steps_per_second": 4.07, | |
| "eval_wer": 1.0, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 23.93617021276596, | |
| "grad_norm": 3.3926920890808105, | |
| "learning_rate": 0.000215856, | |
| "loss": 0.424, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 23.93617021276596, | |
| "eval_cer": 0.30421387441257003, | |
| "eval_loss": 0.8894767761230469, | |
| "eval_runtime": 168.9525, | |
| "eval_samples_per_second": 29.363, | |
| "eval_steps_per_second": 3.676, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 24.20212765957447, | |
| "eval_cer": 0.3013890033976559, | |
| "eval_loss": 0.8863051533699036, | |
| "eval_runtime": 151.1238, | |
| "eval_samples_per_second": 32.827, | |
| "eval_steps_per_second": 4.109, | |
| "eval_wer": 1.0, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 24.46808510638298, | |
| "eval_cer": 0.3112391680733165, | |
| "eval_loss": 0.9255392551422119, | |
| "eval_runtime": 151.1678, | |
| "eval_samples_per_second": 32.818, | |
| "eval_steps_per_second": 4.108, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 24.73404255319149, | |
| "eval_cer": 0.3010635574281958, | |
| "eval_loss": 0.9398043155670166, | |
| "eval_runtime": 151.9435, | |
| "eval_samples_per_second": 32.65, | |
| "eval_steps_per_second": 4.087, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_cer": 0.30710383462137614, | |
| "eval_loss": 0.8763216137886047, | |
| "eval_runtime": 157.5659, | |
| "eval_samples_per_second": 31.485, | |
| "eval_steps_per_second": 3.941, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 25.26595744680851, | |
| "grad_norm": 9.72305679321289, | |
| "learning_rate": 0.00022785599999999997, | |
| "loss": 0.4122, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 25.26595744680851, | |
| "eval_cer": 0.3090174569218018, | |
| "eval_loss": 0.9353150129318237, | |
| "eval_runtime": 152.5497, | |
| "eval_samples_per_second": 32.521, | |
| "eval_steps_per_second": 4.071, | |
| "eval_wer": 1.0, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 25.53191489361702, | |
| "eval_cer": 0.31283602296346763, | |
| "eval_loss": 0.9381543397903442, | |
| "eval_runtime": 171.9179, | |
| "eval_samples_per_second": 28.857, | |
| "eval_steps_per_second": 3.612, | |
| "eval_wer": 1.0, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 25.79787234042553, | |
| "eval_cer": 0.3101673660138944, | |
| "eval_loss": 0.9294571876525879, | |
| "eval_runtime": 151.6573, | |
| "eval_samples_per_second": 32.712, | |
| "eval_steps_per_second": 4.095, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 26.06382978723404, | |
| "eval_cer": 0.3091953673851067, | |
| "eval_loss": 0.9286412596702576, | |
| "eval_runtime": 151.2277, | |
| "eval_samples_per_second": 32.805, | |
| "eval_steps_per_second": 4.106, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 26.329787234042552, | |
| "eval_cer": 0.3013759855588775, | |
| "eval_loss": 0.9140869379043579, | |
| "eval_runtime": 151.0836, | |
| "eval_samples_per_second": 32.836, | |
| "eval_steps_per_second": 4.11, | |
| "eval_wer": 1.0, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 26.595744680851062, | |
| "grad_norm": 4.485568046569824, | |
| "learning_rate": 0.000239856, | |
| "loss": 0.4146, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 26.595744680851062, | |
| "eval_cer": 0.3125756661878995, | |
| "eval_loss": 0.942582905292511, | |
| "eval_runtime": 151.6657, | |
| "eval_samples_per_second": 32.71, | |
| "eval_steps_per_second": 4.095, | |
| "eval_wer": 1.0, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 26.861702127659573, | |
| "eval_cer": 0.3032115008266328, | |
| "eval_loss": 0.8652423620223999, | |
| "eval_runtime": 150.762, | |
| "eval_samples_per_second": 32.906, | |
| "eval_steps_per_second": 4.119, | |
| "eval_wer": 1.0, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 27.127659574468087, | |
| "eval_cer": 0.31053186549968975, | |
| "eval_loss": 0.9288522601127625, | |
| "eval_runtime": 151.4303, | |
| "eval_samples_per_second": 32.761, | |
| "eval_steps_per_second": 4.101, | |
| "eval_wer": 1.0, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 27.393617021276597, | |
| "eval_cer": 0.31025849088534324, | |
| "eval_loss": 0.9458552002906799, | |
| "eval_runtime": 151.6351, | |
| "eval_samples_per_second": 32.717, | |
| "eval_steps_per_second": 4.095, | |
| "eval_wer": 1.0, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 27.659574468085108, | |
| "eval_cer": 0.3123847378858162, | |
| "eval_loss": 0.9137027859687805, | |
| "eval_runtime": 151.0095, | |
| "eval_samples_per_second": 32.852, | |
| "eval_steps_per_second": 4.112, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 27.925531914893618, | |
| "grad_norm": 9.41711711883545, | |
| "learning_rate": 0.000251856, | |
| "loss": 0.416, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 27.925531914893618, | |
| "eval_cer": 0.3099504020342543, | |
| "eval_loss": 0.9305452704429626, | |
| "eval_runtime": 151.4404, | |
| "eval_samples_per_second": 32.759, | |
| "eval_steps_per_second": 4.101, | |
| "eval_wer": 1.0, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 28.19148936170213, | |
| "eval_cer": 0.30706912038463374, | |
| "eval_loss": 0.9589177370071411, | |
| "eval_runtime": 152.6851, | |
| "eval_samples_per_second": 32.492, | |
| "eval_steps_per_second": 4.067, | |
| "eval_wer": 1.0, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 28.45744680851064, | |
| "eval_cer": 0.30605372895991806, | |
| "eval_loss": 0.92755526304245, | |
| "eval_runtime": 151.1956, | |
| "eval_samples_per_second": 32.812, | |
| "eval_steps_per_second": 4.107, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 28.72340425531915, | |
| "eval_cer": 0.3078067979154101, | |
| "eval_loss": 0.9259202480316162, | |
| "eval_runtime": 151.1107, | |
| "eval_samples_per_second": 32.83, | |
| "eval_steps_per_second": 4.11, | |
| "eval_wer": 1.0, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 28.98936170212766, | |
| "eval_cer": 0.3149796270823118, | |
| "eval_loss": 0.9287381768226624, | |
| "eval_runtime": 151.0264, | |
| "eval_samples_per_second": 32.849, | |
| "eval_steps_per_second": 4.112, | |
| "eval_wer": 1.0, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 29.25531914893617, | |
| "grad_norm": 6.876596927642822, | |
| "learning_rate": 0.000263832, | |
| "loss": 0.4078, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 29.25531914893617, | |
| "eval_cer": 0.3165591248540917, | |
| "eval_loss": 0.9346364736557007, | |
| "eval_runtime": 151.4411, | |
| "eval_samples_per_second": 32.759, | |
| "eval_steps_per_second": 4.101, | |
| "eval_wer": 0.9997984277363435, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 29.52127659574468, | |
| "eval_cer": 0.31648969638060687, | |
| "eval_loss": 0.9558972120285034, | |
| "eval_runtime": 151.7115, | |
| "eval_samples_per_second": 32.7, | |
| "eval_steps_per_second": 4.093, | |
| "eval_wer": 1.0, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 29.78723404255319, | |
| "eval_cer": 0.31732717734201765, | |
| "eval_loss": 0.9669720530509949, | |
| "eval_runtime": 151.1358, | |
| "eval_samples_per_second": 32.825, | |
| "eval_steps_per_second": 4.109, | |
| "eval_wer": 1.0, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "step": 11280, | |
| "total_flos": 1.1567283350118728e+19, | |
| "train_loss": 1.009270377192937, | |
| "train_runtime": 42299.7955, | |
| "train_samples_per_second": 8.533, | |
| "train_steps_per_second": 0.267 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 11280, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 400, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1567283350118728e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |