| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 193.21963394342762, | |
| "global_step": 464500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 1.997920133111481e-05, | |
| "loss": 3.456, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 1.995840266222962e-05, | |
| "loss": 1.8719, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.9937603993344426e-05, | |
| "loss": 1.7095, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 1.9916805324459236e-05, | |
| "loss": 1.6155, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_cer": 0.8924, | |
| "eval_gen_len": 19.884, | |
| "eval_loss": 1.4984486103057861, | |
| "eval_runtime": 548.3383, | |
| "eval_samples_per_second": 5.847, | |
| "eval_steps_per_second": 1.463, | |
| "step": 2404 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 1.9896006655574045e-05, | |
| "loss": 1.5555, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1.9875207986688854e-05, | |
| "loss": 1.4646, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 1.9854409317803664e-05, | |
| "loss": 1.4079, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 1.983361064891847e-05, | |
| "loss": 1.3488, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 1.981281198003328e-05, | |
| "loss": 1.3039, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_cer": 0.5947, | |
| "eval_gen_len": 15.5827, | |
| "eval_loss": 1.144504427909851, | |
| "eval_runtime": 282.2519, | |
| "eval_samples_per_second": 11.359, | |
| "eval_steps_per_second": 2.841, | |
| "step": 4808 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.979201331114809e-05, | |
| "loss": 1.2269, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 1.9771214642262898e-05, | |
| "loss": 1.1561, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 1.9750415973377707e-05, | |
| "loss": 1.1288, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 1.9729700499168054e-05, | |
| "loss": 1.0835, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 1.9708901830282864e-05, | |
| "loss": 1.0647, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_cer": 0.4993, | |
| "eval_gen_len": 14.9208, | |
| "eval_loss": 0.9073113799095154, | |
| "eval_runtime": 267.439, | |
| "eval_samples_per_second": 11.988, | |
| "eval_steps_per_second": 2.999, | |
| "step": 7212 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 1.9688103161397673e-05, | |
| "loss": 0.9848, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 1.966730449251248e-05, | |
| "loss": 0.9265, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 1.964650582362729e-05, | |
| "loss": 0.9057, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 1.9625707154742098e-05, | |
| "loss": 0.8799, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 1.9604950083194678e-05, | |
| "loss": 0.8491, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_cer": 0.4462, | |
| "eval_gen_len": 13.549, | |
| "eval_loss": 0.7219734191894531, | |
| "eval_runtime": 263.8796, | |
| "eval_samples_per_second": 12.149, | |
| "eval_steps_per_second": 3.039, | |
| "step": 9616 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 1.9584151414309487e-05, | |
| "loss": 0.7793, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 1.9563352745424296e-05, | |
| "loss": 0.7483, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 1.9542554076539106e-05, | |
| "loss": 0.732, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 1.9521755407653912e-05, | |
| "loss": 0.7186, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 1.9500956738768718e-05, | |
| "loss": 0.7024, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_cer": 0.4716, | |
| "eval_gen_len": 15.7813, | |
| "eval_loss": 0.6081481575965881, | |
| "eval_runtime": 336.6018, | |
| "eval_samples_per_second": 9.525, | |
| "eval_steps_per_second": 2.383, | |
| "step": 12020 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "learning_rate": 1.9480158069883527e-05, | |
| "loss": 0.6114, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 1.945940099833611e-05, | |
| "loss": 0.6193, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "learning_rate": 1.9438602329450916e-05, | |
| "loss": 0.6114, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "learning_rate": 1.9417803660565726e-05, | |
| "loss": 0.5797, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_cer": 0.3726, | |
| "eval_gen_len": 13.8163, | |
| "eval_loss": 0.5052656531333923, | |
| "eval_runtime": 259.2943, | |
| "eval_samples_per_second": 12.364, | |
| "eval_steps_per_second": 3.093, | |
| "step": 14424 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 1.9397004991680535e-05, | |
| "loss": 0.5658, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 1.9376206322795344e-05, | |
| "loss": 0.52, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 1.9355407653910154e-05, | |
| "loss": 0.4992, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 6.66, | |
| "learning_rate": 1.933460898502496e-05, | |
| "loss": 0.5095, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "learning_rate": 1.931381031613977e-05, | |
| "loss": 0.4966, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_cer": 0.3997, | |
| "eval_gen_len": 15.0505, | |
| "eval_loss": 0.4698517620563507, | |
| "eval_runtime": 299.617, | |
| "eval_samples_per_second": 10.7, | |
| "eval_steps_per_second": 2.677, | |
| "step": 16828 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 1.929305324459235e-05, | |
| "loss": 0.4493, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 7.28, | |
| "learning_rate": 1.9272254575707155e-05, | |
| "loss": 0.4288, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "learning_rate": 1.9251455906821964e-05, | |
| "loss": 0.4397, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 7.7, | |
| "learning_rate": 1.9230657237936774e-05, | |
| "loss": 0.423, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 7.9, | |
| "learning_rate": 1.9209900166389354e-05, | |
| "loss": 0.4234, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_cer": 0.3414, | |
| "eval_gen_len": 14.2976, | |
| "eval_loss": 0.41973158717155457, | |
| "eval_runtime": 277.9117, | |
| "eval_samples_per_second": 11.536, | |
| "eval_steps_per_second": 2.886, | |
| "step": 19232 | |
| }, | |
| { | |
| "epoch": 8.11, | |
| "learning_rate": 1.918910149750416e-05, | |
| "loss": 0.377, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "learning_rate": 1.916834442595674e-05, | |
| "loss": 0.3612, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.53, | |
| "learning_rate": 1.914754575707155e-05, | |
| "loss": 0.3725, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "learning_rate": 1.9126747088186358e-05, | |
| "loss": 0.3607, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "learning_rate": 1.9105948419301164e-05, | |
| "loss": 0.3661, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_cer": 0.3568, | |
| "eval_gen_len": 14.5349, | |
| "eval_loss": 0.36951112747192383, | |
| "eval_runtime": 287.7236, | |
| "eval_samples_per_second": 11.143, | |
| "eval_steps_per_second": 2.787, | |
| "step": 21636 | |
| }, | |
| { | |
| "epoch": 9.15, | |
| "learning_rate": 1.9085149750415974e-05, | |
| "loss": 0.3287, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 9.36, | |
| "learning_rate": 1.9064351081530783e-05, | |
| "loss": 0.3185, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "learning_rate": 1.9043552412645592e-05, | |
| "loss": 0.3082, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "learning_rate": 1.90227537437604e-05, | |
| "loss": 0.3177, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 9.98, | |
| "learning_rate": 1.9001996672212978e-05, | |
| "loss": 0.3094, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_cer": 0.3123, | |
| "eval_gen_len": 13.6931, | |
| "eval_loss": 0.35325565934181213, | |
| "eval_runtime": 287.7264, | |
| "eval_samples_per_second": 11.143, | |
| "eval_steps_per_second": 2.787, | |
| "step": 24040 | |
| }, | |
| { | |
| "epoch": 10.19, | |
| "learning_rate": 1.8981198003327787e-05, | |
| "loss": 0.2623, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 1.8960399334442597e-05, | |
| "loss": 0.2666, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 10.61, | |
| "learning_rate": 1.8939600665557406e-05, | |
| "loss": 0.2667, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 10.82, | |
| "learning_rate": 1.8918843594009986e-05, | |
| "loss": 0.2695, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_cer": 0.3036, | |
| "eval_gen_len": 13.8085, | |
| "eval_loss": 0.34151870012283325, | |
| "eval_runtime": 267.5842, | |
| "eval_samples_per_second": 11.981, | |
| "eval_steps_per_second": 2.997, | |
| "step": 26444 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 1.8898086522462566e-05, | |
| "loss": 0.2818, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 11.23, | |
| "learning_rate": 1.8877287853577372e-05, | |
| "loss": 0.24, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 11.44, | |
| "learning_rate": 1.885648918469218e-05, | |
| "loss": 0.2347, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 11.65, | |
| "learning_rate": 1.883569051580699e-05, | |
| "loss": 0.2372, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 1.88148918469218e-05, | |
| "loss": 0.242, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_cer": 0.2954, | |
| "eval_gen_len": 13.5817, | |
| "eval_loss": 0.33137527108192444, | |
| "eval_runtime": 249.8381, | |
| "eval_samples_per_second": 12.832, | |
| "eval_steps_per_second": 3.21, | |
| "step": 28848 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 1.8794093178036606e-05, | |
| "loss": 0.2294, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 12.27, | |
| "learning_rate": 1.8773336106489186e-05, | |
| "loss": 0.2139, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 1.8752537437603995e-05, | |
| "loss": 0.2173, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "learning_rate": 1.8731738768718805e-05, | |
| "loss": 0.214, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 1.871094009983361e-05, | |
| "loss": 0.2123, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_cer": 0.2927, | |
| "eval_gen_len": 13.8079, | |
| "eval_loss": 0.3195803463459015, | |
| "eval_runtime": 264.0497, | |
| "eval_samples_per_second": 12.142, | |
| "eval_steps_per_second": 3.037, | |
| "step": 31252 | |
| }, | |
| { | |
| "epoch": 13.1, | |
| "learning_rate": 1.869014143094842e-05, | |
| "loss": 0.191, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 13.31, | |
| "learning_rate": 1.866934276206323e-05, | |
| "loss": 0.187, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 13.52, | |
| "learning_rate": 1.864854409317804e-05, | |
| "loss": 0.1903, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 13.73, | |
| "learning_rate": 1.8627745424292848e-05, | |
| "loss": 0.1978, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "learning_rate": 1.8606946755407654e-05, | |
| "loss": 0.1954, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_cer": 0.2802, | |
| "eval_gen_len": 13.5215, | |
| "eval_loss": 0.3065420985221863, | |
| "eval_runtime": 255.1945, | |
| "eval_samples_per_second": 12.563, | |
| "eval_steps_per_second": 3.143, | |
| "step": 33656 | |
| }, | |
| { | |
| "epoch": 14.14, | |
| "learning_rate": 1.8586231281198004e-05, | |
| "loss": 0.1658, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 14.35, | |
| "learning_rate": 1.8565432612312814e-05, | |
| "loss": 0.165, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 14.56, | |
| "learning_rate": 1.854463394342762e-05, | |
| "loss": 0.1673, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 14.77, | |
| "learning_rate": 1.852383527454243e-05, | |
| "loss": 0.174, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "learning_rate": 1.850303660565724e-05, | |
| "loss": 0.1734, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_cer": 0.301, | |
| "eval_gen_len": 14.1818, | |
| "eval_loss": 0.3554905652999878, | |
| "eval_runtime": 265.2219, | |
| "eval_samples_per_second": 12.088, | |
| "eval_steps_per_second": 3.024, | |
| "step": 36060 | |
| }, | |
| { | |
| "epoch": 15.18, | |
| "learning_rate": 1.8482237936772048e-05, | |
| "loss": 0.1436, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 15.39, | |
| "learning_rate": 1.8461439267886857e-05, | |
| "loss": 0.1566, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 15.6, | |
| "learning_rate": 1.8440640599001663e-05, | |
| "loss": 0.1548, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 15.81, | |
| "learning_rate": 1.8419883527454246e-05, | |
| "loss": 0.1598, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_cer": 0.2854, | |
| "eval_gen_len": 13.975, | |
| "eval_loss": 0.3254939913749695, | |
| "eval_runtime": 272.7931, | |
| "eval_samples_per_second": 11.752, | |
| "eval_steps_per_second": 2.94, | |
| "step": 38464 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "learning_rate": 1.8399084858569052e-05, | |
| "loss": 0.1543, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 16.22, | |
| "learning_rate": 1.8378286189683862e-05, | |
| "loss": 0.136, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 16.43, | |
| "learning_rate": 1.835748752079867e-05, | |
| "loss": 0.1356, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 16.64, | |
| "learning_rate": 1.833673044925125e-05, | |
| "loss": 0.14, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 16.85, | |
| "learning_rate": 1.8315931780366057e-05, | |
| "loss": 0.1442, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_cer": 0.2748, | |
| "eval_gen_len": 13.4192, | |
| "eval_loss": 0.3075180649757385, | |
| "eval_runtime": 259.5551, | |
| "eval_samples_per_second": 12.352, | |
| "eval_steps_per_second": 3.09, | |
| "step": 40868 | |
| }, | |
| { | |
| "epoch": 17.05, | |
| "learning_rate": 1.8295133111480866e-05, | |
| "loss": 0.1417, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 17.26, | |
| "learning_rate": 1.8274334442595676e-05, | |
| "loss": 0.1241, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 17.47, | |
| "learning_rate": 1.8253577371048256e-05, | |
| "loss": 0.1254, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 17.68, | |
| "learning_rate": 1.823277870216306e-05, | |
| "loss": 0.1315, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 17.89, | |
| "learning_rate": 1.821198003327787e-05, | |
| "loss": 0.1288, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_cer": 0.2763, | |
| "eval_gen_len": 13.5976, | |
| "eval_loss": 0.30095288157463074, | |
| "eval_runtime": 261.6386, | |
| "eval_samples_per_second": 12.254, | |
| "eval_steps_per_second": 3.065, | |
| "step": 43272 | |
| }, | |
| { | |
| "epoch": 18.09, | |
| "learning_rate": 1.819118136439268e-05, | |
| "loss": 0.123, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 18.3, | |
| "learning_rate": 1.817042429284526e-05, | |
| "loss": 0.1178, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 18.51, | |
| "learning_rate": 1.8149625623960066e-05, | |
| "loss": 0.1203, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 18.72, | |
| "learning_rate": 1.8128826955074876e-05, | |
| "loss": 0.1206, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 18.93, | |
| "learning_rate": 1.8108028286189685e-05, | |
| "loss": 0.1249, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_cer": 0.2781, | |
| "eval_gen_len": 13.9969, | |
| "eval_loss": 0.3196047246456146, | |
| "eval_runtime": 267.3702, | |
| "eval_samples_per_second": 11.991, | |
| "eval_steps_per_second": 3.0, | |
| "step": 45676 | |
| }, | |
| { | |
| "epoch": 19.13, | |
| "learning_rate": 1.8087271214642265e-05, | |
| "loss": 0.1067, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 19.34, | |
| "learning_rate": 1.8066514143094845e-05, | |
| "loss": 0.106, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 19.55, | |
| "learning_rate": 1.8045715474209654e-05, | |
| "loss": 0.1083, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 19.76, | |
| "learning_rate": 1.802491680532446e-05, | |
| "loss": 0.1098, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 19.97, | |
| "learning_rate": 1.800411813643927e-05, | |
| "loss": 0.1182, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_cer": 0.2783, | |
| "eval_gen_len": 13.6728, | |
| "eval_loss": 0.3209761679172516, | |
| "eval_runtime": 262.3576, | |
| "eval_samples_per_second": 12.22, | |
| "eval_steps_per_second": 3.057, | |
| "step": 48080 | |
| }, | |
| { | |
| "epoch": 20.17, | |
| "learning_rate": 1.798331946755408e-05, | |
| "loss": 0.0975, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 20.38, | |
| "learning_rate": 1.7962520798668888e-05, | |
| "loss": 0.0975, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 20.59, | |
| "learning_rate": 1.7941722129783698e-05, | |
| "loss": 0.1024, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "learning_rate": 1.7920923460898504e-05, | |
| "loss": 0.1056, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_cer": 0.2707, | |
| "eval_gen_len": 13.5714, | |
| "eval_loss": 0.3157811760902405, | |
| "eval_runtime": 260.7596, | |
| "eval_samples_per_second": 12.295, | |
| "eval_steps_per_second": 3.076, | |
| "step": 50484 | |
| }, | |
| { | |
| "epoch": 21.01, | |
| "learning_rate": 1.7900166389351083e-05, | |
| "loss": 0.1029, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 21.21, | |
| "learning_rate": 1.7879367720465893e-05, | |
| "loss": 0.0943, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 21.42, | |
| "learning_rate": 1.7858569051580702e-05, | |
| "loss": 0.0929, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 21.63, | |
| "learning_rate": 1.7837770382695508e-05, | |
| "loss": 0.0965, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 21.84, | |
| "learning_rate": 1.7816971713810317e-05, | |
| "loss": 0.0916, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_cer": 0.2685, | |
| "eval_gen_len": 13.6404, | |
| "eval_loss": 0.315157949924469, | |
| "eval_runtime": 259.3506, | |
| "eval_samples_per_second": 12.362, | |
| "eval_steps_per_second": 3.092, | |
| "step": 52888 | |
| }, | |
| { | |
| "epoch": 22.05, | |
| "learning_rate": 1.7796173044925127e-05, | |
| "loss": 0.0992, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 22.25, | |
| "learning_rate": 1.7775415973377707e-05, | |
| "loss": 0.0884, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 22.46, | |
| "learning_rate": 1.7754617304492513e-05, | |
| "loss": 0.0922, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 22.67, | |
| "learning_rate": 1.7733818635607322e-05, | |
| "loss": 0.0826, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 22.88, | |
| "learning_rate": 1.771301996672213e-05, | |
| "loss": 0.0934, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_cer": 0.2711, | |
| "eval_gen_len": 13.8575, | |
| "eval_loss": 0.3333517909049988, | |
| "eval_runtime": 261.6177, | |
| "eval_samples_per_second": 12.255, | |
| "eval_steps_per_second": 3.066, | |
| "step": 55292 | |
| }, | |
| { | |
| "epoch": 23.09, | |
| "learning_rate": 1.769222129783694e-05, | |
| "loss": 0.0861, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 23.29, | |
| "learning_rate": 1.7671464226289517e-05, | |
| "loss": 0.081, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 23.5, | |
| "learning_rate": 1.7650665557404327e-05, | |
| "loss": 0.0798, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 23.71, | |
| "learning_rate": 1.7629866888519136e-05, | |
| "loss": 0.0843, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 23.92, | |
| "learning_rate": 1.7609068219633945e-05, | |
| "loss": 0.0941, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_cer": 0.2656, | |
| "eval_gen_len": 13.5583, | |
| "eval_loss": 0.3143361210823059, | |
| "eval_runtime": 261.1394, | |
| "eval_samples_per_second": 12.277, | |
| "eval_steps_per_second": 3.071, | |
| "step": 57696 | |
| }, | |
| { | |
| "epoch": 24.13, | |
| "learning_rate": 1.7588269550748755e-05, | |
| "loss": 0.0809, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 24.33, | |
| "learning_rate": 1.756747088186356e-05, | |
| "loss": 0.0784, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 24.54, | |
| "learning_rate": 1.754671381031614e-05, | |
| "loss": 0.0785, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 24.75, | |
| "learning_rate": 1.752591514143095e-05, | |
| "loss": 0.0782, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 24.96, | |
| "learning_rate": 1.750511647254576e-05, | |
| "loss": 0.0773, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_cer": 0.2641, | |
| "eval_gen_len": 13.5561, | |
| "eval_loss": 0.3231368660926819, | |
| "eval_runtime": 255.8084, | |
| "eval_samples_per_second": 12.533, | |
| "eval_steps_per_second": 3.135, | |
| "step": 60100 | |
| }, | |
| { | |
| "epoch": 25.17, | |
| "learning_rate": 1.7484317803660565e-05, | |
| "loss": 0.0744, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 25.37, | |
| "learning_rate": 1.7463519134775375e-05, | |
| "loss": 0.0719, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 25.58, | |
| "learning_rate": 1.7442762063227955e-05, | |
| "loss": 0.0821, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 25.79, | |
| "learning_rate": 1.7421963394342764e-05, | |
| "loss": 0.0781, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 1.740116472545757e-05, | |
| "loss": 0.0759, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_cer": 0.2668, | |
| "eval_gen_len": 13.7564, | |
| "eval_loss": 0.3243275582790375, | |
| "eval_runtime": 262.8593, | |
| "eval_samples_per_second": 12.197, | |
| "eval_steps_per_second": 3.051, | |
| "step": 62504 | |
| }, | |
| { | |
| "epoch": 26.21, | |
| "learning_rate": 1.738036605657238e-05, | |
| "loss": 0.0664, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 26.41, | |
| "learning_rate": 1.735956738768719e-05, | |
| "loss": 0.0683, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 26.62, | |
| "learning_rate": 1.733881031613977e-05, | |
| "loss": 0.0712, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 26.83, | |
| "learning_rate": 1.7318011647254578e-05, | |
| "loss": 0.077, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_cer": 0.267, | |
| "eval_gen_len": 13.7558, | |
| "eval_loss": 0.33405444025993347, | |
| "eval_runtime": 259.6084, | |
| "eval_samples_per_second": 12.349, | |
| "eval_steps_per_second": 3.089, | |
| "step": 64908 | |
| }, | |
| { | |
| "epoch": 27.04, | |
| "learning_rate": 1.7297212978369387e-05, | |
| "loss": 0.076, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 27.25, | |
| "learning_rate": 1.7276414309484197e-05, | |
| "loss": 0.0595, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 27.45, | |
| "learning_rate": 1.7255615640599003e-05, | |
| "loss": 0.0608, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 27.66, | |
| "learning_rate": 1.7234858569051583e-05, | |
| "loss": 0.0705, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 27.87, | |
| "learning_rate": 1.7214059900166392e-05, | |
| "loss": 0.0743, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_cer": 0.2796, | |
| "eval_gen_len": 14.1085, | |
| "eval_loss": 0.3349040448665619, | |
| "eval_runtime": 282.8059, | |
| "eval_samples_per_second": 11.336, | |
| "eval_steps_per_second": 2.836, | |
| "step": 67312 | |
| }, | |
| { | |
| "epoch": 28.08, | |
| "learning_rate": 1.71932612312812e-05, | |
| "loss": 0.0674, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 28.29, | |
| "learning_rate": 1.7172462562396007e-05, | |
| "loss": 0.0621, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 28.49, | |
| "learning_rate": 1.7151663893510817e-05, | |
| "loss": 0.0661, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 28.7, | |
| "learning_rate": 1.7130865224625626e-05, | |
| "loss": 0.0611, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 28.91, | |
| "learning_rate": 1.7110108153078206e-05, | |
| "loss": 0.0662, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_cer": 0.2623, | |
| "eval_gen_len": 13.5396, | |
| "eval_loss": 0.3251936137676239, | |
| "eval_runtime": 258.4167, | |
| "eval_samples_per_second": 12.406, | |
| "eval_steps_per_second": 3.104, | |
| "step": 69716 | |
| }, | |
| { | |
| "epoch": 29.12, | |
| "learning_rate": 1.7089309484193012e-05, | |
| "loss": 0.0652, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 29.33, | |
| "learning_rate": 1.706851081530782e-05, | |
| "loss": 0.0557, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 29.53, | |
| "learning_rate": 1.704771214642263e-05, | |
| "loss": 0.0584, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 29.74, | |
| "learning_rate": 1.702691347753744e-05, | |
| "loss": 0.0627, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 29.95, | |
| "learning_rate": 1.700611480865225e-05, | |
| "loss": 0.0685, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_cer": 0.2643, | |
| "eval_gen_len": 13.6528, | |
| "eval_loss": 0.33093786239624023, | |
| "eval_runtime": 269.9487, | |
| "eval_samples_per_second": 11.876, | |
| "eval_steps_per_second": 2.971, | |
| "step": 72120 | |
| }, | |
| { | |
| "epoch": 30.16, | |
| "learning_rate": 1.6985357737104826e-05, | |
| "loss": 0.0592, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 30.37, | |
| "learning_rate": 1.6964559068219635e-05, | |
| "loss": 0.0564, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 30.57, | |
| "learning_rate": 1.6943760399334445e-05, | |
| "loss": 0.0623, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 30.78, | |
| "learning_rate": 1.692296173044925e-05, | |
| "loss": 0.0599, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 30.99, | |
| "learning_rate": 1.690220465890183e-05, | |
| "loss": 0.0619, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_cer": 0.266, | |
| "eval_gen_len": 13.7171, | |
| "eval_loss": 0.3532868027687073, | |
| "eval_runtime": 257.8089, | |
| "eval_samples_per_second": 12.436, | |
| "eval_steps_per_second": 3.111, | |
| "step": 74524 | |
| }, | |
| { | |
| "epoch": 31.2, | |
| "learning_rate": 1.688140599001664e-05, | |
| "loss": 0.0499, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 31.41, | |
| "learning_rate": 1.686060732113145e-05, | |
| "loss": 0.0592, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 31.61, | |
| "learning_rate": 1.6839808652246255e-05, | |
| "loss": 0.0555, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 31.82, | |
| "learning_rate": 1.6819009983361064e-05, | |
| "loss": 0.0602, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_cer": 0.2661, | |
| "eval_gen_len": 13.8924, | |
| "eval_loss": 0.34633541107177734, | |
| "eval_runtime": 271.365, | |
| "eval_samples_per_second": 11.814, | |
| "eval_steps_per_second": 2.955, | |
| "step": 76928 | |
| }, | |
| { | |
| "epoch": 32.03, | |
| "learning_rate": 1.6798211314475874e-05, | |
| "loss": 0.0575, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 32.24, | |
| "learning_rate": 1.6777412645590683e-05, | |
| "loss": 0.0506, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 32.45, | |
| "learning_rate": 1.6756613976705493e-05, | |
| "loss": 0.0566, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 32.65, | |
| "learning_rate": 1.6735856905158072e-05, | |
| "loss": 0.0532, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 32.86, | |
| "learning_rate": 1.6715058236272882e-05, | |
| "loss": 0.0568, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_cer": 0.2645, | |
| "eval_gen_len": 13.5443, | |
| "eval_loss": 0.3539523482322693, | |
| "eval_runtime": 256.805, | |
| "eval_samples_per_second": 12.484, | |
| "eval_steps_per_second": 3.123, | |
| "step": 79332 | |
| }, | |
| { | |
| "epoch": 33.07, | |
| "learning_rate": 1.669425956738769e-05, | |
| "loss": 0.057, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 33.28, | |
| "learning_rate": 1.6673460898502497e-05, | |
| "loss": 0.0467, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 33.49, | |
| "learning_rate": 1.6652703826955077e-05, | |
| "loss": 0.0514, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 33.69, | |
| "learning_rate": 1.6631905158069886e-05, | |
| "loss": 0.0522, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 33.9, | |
| "learning_rate": 1.6611106489184692e-05, | |
| "loss": 0.0528, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_cer": 0.2621, | |
| "eval_gen_len": 13.4984, | |
| "eval_loss": 0.3489411473274231, | |
| "eval_runtime": 254.1092, | |
| "eval_samples_per_second": 12.617, | |
| "eval_steps_per_second": 3.156, | |
| "step": 81736 | |
| }, | |
| { | |
| "epoch": 34.11, | |
| "learning_rate": 1.6590307820299502e-05, | |
| "loss": 0.05, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 34.32, | |
| "learning_rate": 1.656955074875208e-05, | |
| "loss": 0.0477, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 34.53, | |
| "learning_rate": 1.654875207986689e-05, | |
| "loss": 0.0567, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 34.73, | |
| "learning_rate": 1.6527953410981697e-05, | |
| "loss": 0.0477, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 34.94, | |
| "learning_rate": 1.6507154742096506e-05, | |
| "loss": 0.0544, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_cer": 0.2662, | |
| "eval_gen_len": 13.7679, | |
| "eval_loss": 0.3497730493545532, | |
| "eval_runtime": 268.9078, | |
| "eval_samples_per_second": 11.922, | |
| "eval_steps_per_second": 2.982, | |
| "step": 84140 | |
| }, | |
| { | |
| "epoch": 35.15, | |
| "learning_rate": 1.6486356073211316e-05, | |
| "loss": 0.0433, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 35.36, | |
| "learning_rate": 1.6465599001663896e-05, | |
| "loss": 0.0426, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 35.57, | |
| "learning_rate": 1.64448003327787e-05, | |
| "loss": 0.0477, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 35.77, | |
| "learning_rate": 1.642400166389351e-05, | |
| "loss": 0.0497, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 35.98, | |
| "learning_rate": 1.640320299500832e-05, | |
| "loss": 0.0571, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_cer": 0.2661, | |
| "eval_gen_len": 13.8001, | |
| "eval_loss": 0.35858893394470215, | |
| "eval_runtime": 266.9991, | |
| "eval_samples_per_second": 12.008, | |
| "eval_steps_per_second": 3.004, | |
| "step": 86544 | |
| }, | |
| { | |
| "epoch": 36.19, | |
| "learning_rate": 1.63824459234609e-05, | |
| "loss": 0.0406, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 36.4, | |
| "learning_rate": 1.6361647254575706e-05, | |
| "loss": 0.0431, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 36.61, | |
| "learning_rate": 1.6340848585690516e-05, | |
| "loss": 0.0507, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 36.81, | |
| "learning_rate": 1.6320049916805325e-05, | |
| "loss": 0.0492, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_cer": 0.2564, | |
| "eval_gen_len": 13.5415, | |
| "eval_loss": 0.35140517354011536, | |
| "eval_runtime": 259.4147, | |
| "eval_samples_per_second": 12.359, | |
| "eval_steps_per_second": 3.092, | |
| "step": 88948 | |
| }, | |
| { | |
| "epoch": 37.02, | |
| "learning_rate": 1.6299251247920134e-05, | |
| "loss": 0.0515, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 37.23, | |
| "learning_rate": 1.6278452579034944e-05, | |
| "loss": 0.0421, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 37.44, | |
| "learning_rate": 1.6257695507487523e-05, | |
| "loss": 0.0432, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 37.65, | |
| "learning_rate": 1.6236896838602333e-05, | |
| "loss": 0.0449, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 37.85, | |
| "learning_rate": 1.621609816971714e-05, | |
| "loss": 0.0484, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_cer": 0.2639, | |
| "eval_gen_len": 13.8534, | |
| "eval_loss": 0.3746128976345062, | |
| "eval_runtime": 254.0333, | |
| "eval_samples_per_second": 12.62, | |
| "eval_steps_per_second": 3.157, | |
| "step": 91352 | |
| }, | |
| { | |
| "epoch": 38.06, | |
| "learning_rate": 1.6195299500831948e-05, | |
| "loss": 0.0481, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 38.27, | |
| "learning_rate": 1.6174500831946758e-05, | |
| "loss": 0.0402, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 38.48, | |
| "learning_rate": 1.6153743760399337e-05, | |
| "loss": 0.0395, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 38.69, | |
| "learning_rate": 1.6132945091514143e-05, | |
| "loss": 0.0456, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 38.89, | |
| "learning_rate": 1.6112146422628953e-05, | |
| "loss": 0.0451, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_cer": 0.2627, | |
| "eval_gen_len": 13.7527, | |
| "eval_loss": 0.35255342721939087, | |
| "eval_runtime": 264.6452, | |
| "eval_samples_per_second": 12.114, | |
| "eval_steps_per_second": 3.03, | |
| "step": 93756 | |
| }, | |
| { | |
| "epoch": 39.1, | |
| "learning_rate": 1.6091347753743762e-05, | |
| "loss": 0.0363, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 39.31, | |
| "learning_rate": 1.6070590682196342e-05, | |
| "loss": 0.0382, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 39.52, | |
| "learning_rate": 1.6049792013311148e-05, | |
| "loss": 0.0448, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 39.73, | |
| "learning_rate": 1.6028993344425957e-05, | |
| "loss": 0.0399, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 39.93, | |
| "learning_rate": 1.6008194675540767e-05, | |
| "loss": 0.045, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_cer": 0.2583, | |
| "eval_gen_len": 13.6694, | |
| "eval_loss": 0.3510436713695526, | |
| "eval_runtime": 264.9384, | |
| "eval_samples_per_second": 12.101, | |
| "eval_steps_per_second": 3.027, | |
| "step": 96160 | |
| }, | |
| { | |
| "epoch": 40.14, | |
| "learning_rate": 1.5987437603993347e-05, | |
| "loss": 0.0393, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 40.35, | |
| "learning_rate": 1.5966638935108153e-05, | |
| "loss": 0.0338, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 40.56, | |
| "learning_rate": 1.5945840266222962e-05, | |
| "loss": 0.0398, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 40.77, | |
| "learning_rate": 1.5925083194675542e-05, | |
| "loss": 0.0425, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 40.97, | |
| "learning_rate": 1.590428452579035e-05, | |
| "loss": 0.0455, | |
| "step": 98500 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_cer": 0.2619, | |
| "eval_gen_len": 13.7012, | |
| "eval_loss": 0.3675419092178345, | |
| "eval_runtime": 255.341, | |
| "eval_samples_per_second": 12.556, | |
| "eval_steps_per_second": 3.141, | |
| "step": 98564 | |
| }, | |
| { | |
| "epoch": 41.18, | |
| "learning_rate": 1.5883485856905157e-05, | |
| "loss": 0.0372, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 41.39, | |
| "learning_rate": 1.5862687188019967e-05, | |
| "loss": 0.0378, | |
| "step": 99500 | |
| }, | |
| { | |
| "epoch": 41.6, | |
| "learning_rate": 1.5841888519134776e-05, | |
| "loss": 0.0369, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 41.81, | |
| "learning_rate": 1.5821089850249585e-05, | |
| "loss": 0.0452, | |
| "step": 100500 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_cer": 0.2672, | |
| "eval_gen_len": 13.8653, | |
| "eval_loss": 0.3757382035255432, | |
| "eval_runtime": 268.6077, | |
| "eval_samples_per_second": 11.936, | |
| "eval_steps_per_second": 2.986, | |
| "step": 100968 | |
| }, | |
| { | |
| "epoch": 42.01, | |
| "learning_rate": 1.5800291181364395e-05, | |
| "loss": 0.046, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 42.22, | |
| "learning_rate": 1.57794925124792e-05, | |
| "loss": 0.0319, | |
| "step": 101500 | |
| }, | |
| { | |
| "epoch": 42.43, | |
| "learning_rate": 1.5758735440931784e-05, | |
| "loss": 0.0383, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 42.64, | |
| "learning_rate": 1.573793677204659e-05, | |
| "loss": 0.0376, | |
| "step": 102500 | |
| }, | |
| { | |
| "epoch": 42.85, | |
| "learning_rate": 1.57171381031614e-05, | |
| "loss": 0.0403, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_cer": 0.2583, | |
| "eval_gen_len": 13.6532, | |
| "eval_loss": 0.3421362340450287, | |
| "eval_runtime": 257.7664, | |
| "eval_samples_per_second": 12.438, | |
| "eval_steps_per_second": 3.111, | |
| "step": 103372 | |
| }, | |
| { | |
| "epoch": 43.05, | |
| "learning_rate": 1.5696339434276205e-05, | |
| "loss": 0.0419, | |
| "step": 103500 | |
| }, | |
| { | |
| "epoch": 43.26, | |
| "learning_rate": 1.5675540765391015e-05, | |
| "loss": 0.0322, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 43.47, | |
| "learning_rate": 1.5654783693843594e-05, | |
| "loss": 0.0396, | |
| "step": 104500 | |
| }, | |
| { | |
| "epoch": 43.68, | |
| "learning_rate": 1.5633985024958404e-05, | |
| "loss": 0.0386, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 43.89, | |
| "learning_rate": 1.5613186356073213e-05, | |
| "loss": 0.0372, | |
| "step": 105500 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_cer": 0.2553, | |
| "eval_gen_len": 13.7679, | |
| "eval_loss": 0.37172210216522217, | |
| "eval_runtime": 266.5505, | |
| "eval_samples_per_second": 12.028, | |
| "eval_steps_per_second": 3.009, | |
| "step": 105776 | |
| }, | |
| { | |
| "epoch": 44.09, | |
| "learning_rate": 1.5592387687188023e-05, | |
| "loss": 0.0381, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 44.3, | |
| "learning_rate": 1.55716306156406e-05, | |
| "loss": 0.0343, | |
| "step": 106500 | |
| }, | |
| { | |
| "epoch": 44.51, | |
| "learning_rate": 1.555083194675541e-05, | |
| "loss": 0.0348, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 44.72, | |
| "learning_rate": 1.5530033277870218e-05, | |
| "loss": 0.0412, | |
| "step": 107500 | |
| }, | |
| { | |
| "epoch": 44.93, | |
| "learning_rate": 1.5509234608985027e-05, | |
| "loss": 0.041, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_cer": 0.258, | |
| "eval_gen_len": 13.7246, | |
| "eval_loss": 0.367112934589386, | |
| "eval_runtime": 255.2355, | |
| "eval_samples_per_second": 12.561, | |
| "eval_steps_per_second": 3.142, | |
| "step": 108180 | |
| }, | |
| { | |
| "epoch": 45.13, | |
| "learning_rate": 1.5488477537437604e-05, | |
| "loss": 0.0358, | |
| "step": 108500 | |
| }, | |
| { | |
| "epoch": 45.34, | |
| "learning_rate": 1.5467678868552413e-05, | |
| "loss": 0.0347, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 45.55, | |
| "learning_rate": 1.5446880199667222e-05, | |
| "loss": 0.0372, | |
| "step": 109500 | |
| }, | |
| { | |
| "epoch": 45.76, | |
| "learning_rate": 1.5426081530782032e-05, | |
| "loss": 0.0369, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 45.97, | |
| "learning_rate": 1.540528286189684e-05, | |
| "loss": 0.0372, | |
| "step": 110500 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_cer": 0.2596, | |
| "eval_gen_len": 13.8244, | |
| "eval_loss": 0.36525318026542664, | |
| "eval_runtime": 258.7606, | |
| "eval_samples_per_second": 12.39, | |
| "eval_steps_per_second": 3.099, | |
| "step": 110584 | |
| }, | |
| { | |
| "epoch": 46.17, | |
| "learning_rate": 1.5384525790349418e-05, | |
| "loss": 0.0276, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 46.38, | |
| "learning_rate": 1.5363727121464227e-05, | |
| "loss": 0.0311, | |
| "step": 111500 | |
| }, | |
| { | |
| "epoch": 46.59, | |
| "learning_rate": 1.5342928452579036e-05, | |
| "loss": 0.0343, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 46.8, | |
| "learning_rate": 1.5322129783693846e-05, | |
| "loss": 0.0418, | |
| "step": 112500 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_cer": 0.2573, | |
| "eval_gen_len": 13.7427, | |
| "eval_loss": 0.3767205476760864, | |
| "eval_runtime": 259.1923, | |
| "eval_samples_per_second": 12.369, | |
| "eval_steps_per_second": 3.094, | |
| "step": 112988 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "learning_rate": 1.5301372712146422e-05, | |
| "loss": 0.0387, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 47.21, | |
| "learning_rate": 1.528057404326123e-05, | |
| "loss": 0.0288, | |
| "step": 113500 | |
| }, | |
| { | |
| "epoch": 47.42, | |
| "learning_rate": 1.525977537437604e-05, | |
| "loss": 0.0316, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 47.63, | |
| "learning_rate": 1.5238976705490849e-05, | |
| "loss": 0.0302, | |
| "step": 114500 | |
| }, | |
| { | |
| "epoch": 47.84, | |
| "learning_rate": 1.521821963394343e-05, | |
| "loss": 0.036, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_cer": 0.2645, | |
| "eval_gen_len": 13.9616, | |
| "eval_loss": 0.3853040933609009, | |
| "eval_runtime": 258.5981, | |
| "eval_samples_per_second": 12.398, | |
| "eval_steps_per_second": 3.101, | |
| "step": 115392 | |
| }, | |
| { | |
| "epoch": 48.04, | |
| "learning_rate": 1.5197420965058238e-05, | |
| "loss": 0.0414, | |
| "step": 115500 | |
| }, | |
| { | |
| "epoch": 48.25, | |
| "learning_rate": 1.5176622296173047e-05, | |
| "loss": 0.0302, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 48.46, | |
| "learning_rate": 1.5155823627287855e-05, | |
| "loss": 0.0336, | |
| "step": 116500 | |
| }, | |
| { | |
| "epoch": 48.67, | |
| "learning_rate": 1.5135066555740435e-05, | |
| "loss": 0.0346, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 48.88, | |
| "learning_rate": 1.5114267886855242e-05, | |
| "loss": 0.0354, | |
| "step": 117500 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_cer": 0.2571, | |
| "eval_gen_len": 13.3799, | |
| "eval_loss": 0.3713897466659546, | |
| "eval_runtime": 256.4932, | |
| "eval_samples_per_second": 12.499, | |
| "eval_steps_per_second": 3.127, | |
| "step": 117796 | |
| }, | |
| { | |
| "epoch": 49.08, | |
| "learning_rate": 1.5093469217970052e-05, | |
| "loss": 0.0285, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 49.29, | |
| "learning_rate": 1.507267054908486e-05, | |
| "loss": 0.0311, | |
| "step": 118500 | |
| }, | |
| { | |
| "epoch": 49.5, | |
| "learning_rate": 1.5051871880199669e-05, | |
| "loss": 0.0297, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 49.71, | |
| "learning_rate": 1.5031114808652247e-05, | |
| "loss": 0.0327, | |
| "step": 119500 | |
| }, | |
| { | |
| "epoch": 49.92, | |
| "learning_rate": 1.5010316139767056e-05, | |
| "loss": 0.0336, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_cer": 0.2592, | |
| "eval_gen_len": 13.7667, | |
| "eval_loss": 0.38062140345573425, | |
| "eval_runtime": 257.9154, | |
| "eval_samples_per_second": 12.43, | |
| "eval_steps_per_second": 3.11, | |
| "step": 120200 | |
| }, | |
| { | |
| "epoch": 50.12, | |
| "learning_rate": 1.4989517470881864e-05, | |
| "loss": 0.0291, | |
| "step": 120500 | |
| }, | |
| { | |
| "epoch": 50.33, | |
| "learning_rate": 1.4968718801996673e-05, | |
| "loss": 0.0286, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 50.54, | |
| "learning_rate": 1.4947920133111483e-05, | |
| "loss": 0.0282, | |
| "step": 121500 | |
| }, | |
| { | |
| "epoch": 50.75, | |
| "learning_rate": 1.4927163061564061e-05, | |
| "loss": 0.0299, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 50.96, | |
| "learning_rate": 1.4906364392678869e-05, | |
| "loss": 0.0367, | |
| "step": 122500 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_cer": 0.2567, | |
| "eval_gen_len": 13.7402, | |
| "eval_loss": 0.36996766924858093, | |
| "eval_runtime": 265.273, | |
| "eval_samples_per_second": 12.086, | |
| "eval_steps_per_second": 3.023, | |
| "step": 122604 | |
| }, | |
| { | |
| "epoch": 51.16, | |
| "learning_rate": 1.4885607321131449e-05, | |
| "loss": 0.0269, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 51.37, | |
| "learning_rate": 1.4864808652246256e-05, | |
| "loss": 0.0288, | |
| "step": 123500 | |
| }, | |
| { | |
| "epoch": 51.58, | |
| "learning_rate": 1.4844009983361066e-05, | |
| "loss": 0.0306, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 51.79, | |
| "learning_rate": 1.4823211314475873e-05, | |
| "loss": 0.0352, | |
| "step": 124500 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "learning_rate": 1.4802412645590683e-05, | |
| "loss": 0.0278, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_cer": 0.2552, | |
| "eval_gen_len": 13.6507, | |
| "eval_loss": 0.37308937311172485, | |
| "eval_runtime": 264.0574, | |
| "eval_samples_per_second": 12.141, | |
| "eval_steps_per_second": 3.037, | |
| "step": 125008 | |
| }, | |
| { | |
| "epoch": 52.2, | |
| "learning_rate": 1.478161397670549e-05, | |
| "loss": 0.0245, | |
| "step": 125500 | |
| }, | |
| { | |
| "epoch": 52.41, | |
| "learning_rate": 1.47608153078203e-05, | |
| "loss": 0.0256, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 52.62, | |
| "learning_rate": 1.4740016638935109e-05, | |
| "loss": 0.0303, | |
| "step": 126500 | |
| }, | |
| { | |
| "epoch": 52.83, | |
| "learning_rate": 1.4719259567387689e-05, | |
| "loss": 0.0318, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_cer": 0.2558, | |
| "eval_gen_len": 13.4878, | |
| "eval_loss": 0.3711611330509186, | |
| "eval_runtime": 256.1423, | |
| "eval_samples_per_second": 12.516, | |
| "eval_steps_per_second": 3.131, | |
| "step": 127412 | |
| }, | |
| { | |
| "epoch": 53.04, | |
| "learning_rate": 1.4698460898502498e-05, | |
| "loss": 0.0308, | |
| "step": 127500 | |
| }, | |
| { | |
| "epoch": 53.24, | |
| "learning_rate": 1.4677662229617306e-05, | |
| "loss": 0.0258, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 53.45, | |
| "learning_rate": 1.4656863560732115e-05, | |
| "loss": 0.0268, | |
| "step": 128500 | |
| }, | |
| { | |
| "epoch": 53.66, | |
| "learning_rate": 1.4636064891846925e-05, | |
| "loss": 0.0306, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 53.87, | |
| "learning_rate": 1.4615307820299503e-05, | |
| "loss": 0.0307, | |
| "step": 129500 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_cer": 0.2545, | |
| "eval_gen_len": 13.6042, | |
| "eval_loss": 0.3638122081756592, | |
| "eval_runtime": 259.1147, | |
| "eval_samples_per_second": 12.373, | |
| "eval_steps_per_second": 3.095, | |
| "step": 129816 | |
| }, | |
| { | |
| "epoch": 54.08, | |
| "learning_rate": 1.459450915141431e-05, | |
| "loss": 0.0246, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 54.28, | |
| "learning_rate": 1.457371048252912e-05, | |
| "loss": 0.0206, | |
| "step": 130500 | |
| }, | |
| { | |
| "epoch": 54.49, | |
| "learning_rate": 1.455291181364393e-05, | |
| "loss": 0.0298, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 54.7, | |
| "learning_rate": 1.4532196339434276e-05, | |
| "loss": 0.0284, | |
| "step": 131500 | |
| }, | |
| { | |
| "epoch": 54.91, | |
| "learning_rate": 1.4511397670549086e-05, | |
| "loss": 0.0277, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_cer": 0.2574, | |
| "eval_gen_len": 13.8247, | |
| "eval_loss": 0.3866593539714813, | |
| "eval_runtime": 266.2003, | |
| "eval_samples_per_second": 12.044, | |
| "eval_steps_per_second": 3.013, | |
| "step": 132220 | |
| }, | |
| { | |
| "epoch": 55.12, | |
| "learning_rate": 1.4490599001663895e-05, | |
| "loss": 0.0258, | |
| "step": 132500 | |
| }, | |
| { | |
| "epoch": 55.32, | |
| "learning_rate": 1.4469800332778703e-05, | |
| "loss": 0.0264, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 55.53, | |
| "learning_rate": 1.4449001663893512e-05, | |
| "loss": 0.0273, | |
| "step": 133500 | |
| }, | |
| { | |
| "epoch": 55.74, | |
| "learning_rate": 1.442820299500832e-05, | |
| "loss": 0.033, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 55.95, | |
| "learning_rate": 1.4407404326123129e-05, | |
| "loss": 0.0289, | |
| "step": 134500 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_cer": 0.26, | |
| "eval_gen_len": 13.9024, | |
| "eval_loss": 0.38215455412864685, | |
| "eval_runtime": 268.768, | |
| "eval_samples_per_second": 11.929, | |
| "eval_steps_per_second": 2.984, | |
| "step": 134624 | |
| }, | |
| { | |
| "epoch": 56.16, | |
| "learning_rate": 1.4386605657237937e-05, | |
| "loss": 0.0263, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 56.36, | |
| "learning_rate": 1.4365848585690517e-05, | |
| "loss": 0.0227, | |
| "step": 135500 | |
| }, | |
| { | |
| "epoch": 56.57, | |
| "learning_rate": 1.4345049916805324e-05, | |
| "loss": 0.0276, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 56.78, | |
| "learning_rate": 1.4324251247920134e-05, | |
| "loss": 0.0277, | |
| "step": 136500 | |
| }, | |
| { | |
| "epoch": 56.99, | |
| "learning_rate": 1.4303494176372714e-05, | |
| "loss": 0.0259, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_cer": 0.2541, | |
| "eval_gen_len": 13.6722, | |
| "eval_loss": 0.3895968198776245, | |
| "eval_runtime": 256.4716, | |
| "eval_samples_per_second": 12.5, | |
| "eval_steps_per_second": 3.127, | |
| "step": 137028 | |
| }, | |
| { | |
| "epoch": 57.2, | |
| "learning_rate": 1.4282695507487523e-05, | |
| "loss": 0.0202, | |
| "step": 137500 | |
| }, | |
| { | |
| "epoch": 57.4, | |
| "learning_rate": 1.4261896838602332e-05, | |
| "loss": 0.0236, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 57.61, | |
| "learning_rate": 1.424109816971714e-05, | |
| "loss": 0.0288, | |
| "step": 138500 | |
| }, | |
| { | |
| "epoch": 57.82, | |
| "learning_rate": 1.422029950083195e-05, | |
| "loss": 0.0277, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_cer": 0.2584, | |
| "eval_gen_len": 13.7208, | |
| "eval_loss": 0.38816508650779724, | |
| "eval_runtime": 264.0106, | |
| "eval_samples_per_second": 12.143, | |
| "eval_steps_per_second": 3.038, | |
| "step": 139432 | |
| }, | |
| { | |
| "epoch": 58.03, | |
| "learning_rate": 1.4199500831946757e-05, | |
| "loss": 0.0297, | |
| "step": 139500 | |
| }, | |
| { | |
| "epoch": 58.24, | |
| "learning_rate": 1.4178702163061566e-05, | |
| "loss": 0.0206, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 58.44, | |
| "learning_rate": 1.4157945091514145e-05, | |
| "loss": 0.0245, | |
| "step": 140500 | |
| }, | |
| { | |
| "epoch": 58.65, | |
| "learning_rate": 1.4137146422628954e-05, | |
| "loss": 0.0264, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 58.86, | |
| "learning_rate": 1.4116347753743762e-05, | |
| "loss": 0.0289, | |
| "step": 141500 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_cer": 0.2587, | |
| "eval_gen_len": 13.7096, | |
| "eval_loss": 0.39096423983573914, | |
| "eval_runtime": 260.1849, | |
| "eval_samples_per_second": 12.322, | |
| "eval_steps_per_second": 3.082, | |
| "step": 141836 | |
| }, | |
| { | |
| "epoch": 59.07, | |
| "learning_rate": 1.4095549084858571e-05, | |
| "loss": 0.0243, | |
| "step": 142000 | |
| }, | |
| { | |
| "epoch": 59.28, | |
| "learning_rate": 1.4074750415973379e-05, | |
| "loss": 0.0218, | |
| "step": 142500 | |
| }, | |
| { | |
| "epoch": 59.48, | |
| "learning_rate": 1.4053951747088188e-05, | |
| "loss": 0.0241, | |
| "step": 143000 | |
| }, | |
| { | |
| "epoch": 59.69, | |
| "learning_rate": 1.4033194675540766e-05, | |
| "loss": 0.0242, | |
| "step": 143500 | |
| }, | |
| { | |
| "epoch": 59.9, | |
| "learning_rate": 1.4012396006655576e-05, | |
| "loss": 0.0252, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_cer": 0.2515, | |
| "eval_gen_len": 13.6971, | |
| "eval_loss": 0.38846734166145325, | |
| "eval_runtime": 258.3853, | |
| "eval_samples_per_second": 12.408, | |
| "eval_steps_per_second": 3.104, | |
| "step": 144240 | |
| }, | |
| { | |
| "epoch": 60.11, | |
| "learning_rate": 1.3991597337770383e-05, | |
| "loss": 0.0259, | |
| "step": 144500 | |
| }, | |
| { | |
| "epoch": 60.32, | |
| "learning_rate": 1.3970798668885193e-05, | |
| "loss": 0.0208, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 60.52, | |
| "learning_rate": 1.3950000000000002e-05, | |
| "loss": 0.0252, | |
| "step": 145500 | |
| }, | |
| { | |
| "epoch": 60.73, | |
| "learning_rate": 1.392920133111481e-05, | |
| "loss": 0.0253, | |
| "step": 146000 | |
| }, | |
| { | |
| "epoch": 60.94, | |
| "learning_rate": 1.3908402662229619e-05, | |
| "loss": 0.0265, | |
| "step": 146500 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_cer": 0.2569, | |
| "eval_gen_len": 13.85, | |
| "eval_loss": 0.3915986716747284, | |
| "eval_runtime": 271.1284, | |
| "eval_samples_per_second": 11.825, | |
| "eval_steps_per_second": 2.958, | |
| "step": 146644 | |
| }, | |
| { | |
| "epoch": 61.15, | |
| "learning_rate": 1.3887603993344427e-05, | |
| "loss": 0.0232, | |
| "step": 147000 | |
| }, | |
| { | |
| "epoch": 61.36, | |
| "learning_rate": 1.3866846921797007e-05, | |
| "loss": 0.0233, | |
| "step": 147500 | |
| }, | |
| { | |
| "epoch": 61.56, | |
| "learning_rate": 1.3846048252911814e-05, | |
| "loss": 0.0224, | |
| "step": 148000 | |
| }, | |
| { | |
| "epoch": 61.77, | |
| "learning_rate": 1.3825249584026624e-05, | |
| "loss": 0.0229, | |
| "step": 148500 | |
| }, | |
| { | |
| "epoch": 61.98, | |
| "learning_rate": 1.3804450915141431e-05, | |
| "loss": 0.0229, | |
| "step": 149000 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_cer": 0.2565, | |
| "eval_gen_len": 13.8206, | |
| "eval_loss": 0.3992536962032318, | |
| "eval_runtime": 268.8678, | |
| "eval_samples_per_second": 11.924, | |
| "eval_steps_per_second": 2.983, | |
| "step": 149048 | |
| }, | |
| { | |
| "epoch": 62.19, | |
| "learning_rate": 1.378365224625624e-05, | |
| "loss": 0.0199, | |
| "step": 149500 | |
| }, | |
| { | |
| "epoch": 62.4, | |
| "learning_rate": 1.3762895174708819e-05, | |
| "loss": 0.0239, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 62.6, | |
| "learning_rate": 1.3742096505823628e-05, | |
| "loss": 0.0261, | |
| "step": 150500 | |
| }, | |
| { | |
| "epoch": 62.81, | |
| "learning_rate": 1.3721297836938436e-05, | |
| "loss": 0.0225, | |
| "step": 151000 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_cer": 0.2507, | |
| "eval_gen_len": 13.6354, | |
| "eval_loss": 0.3880002200603485, | |
| "eval_runtime": 264.7668, | |
| "eval_samples_per_second": 12.109, | |
| "eval_steps_per_second": 3.029, | |
| "step": 151452 | |
| }, | |
| { | |
| "epoch": 63.02, | |
| "learning_rate": 1.3700499168053245e-05, | |
| "loss": 0.0268, | |
| "step": 151500 | |
| }, | |
| { | |
| "epoch": 63.23, | |
| "learning_rate": 1.3679742096505825e-05, | |
| "loss": 0.0188, | |
| "step": 152000 | |
| }, | |
| { | |
| "epoch": 63.44, | |
| "learning_rate": 1.3658985024958405e-05, | |
| "loss": 0.022, | |
| "step": 152500 | |
| }, | |
| { | |
| "epoch": 63.64, | |
| "learning_rate": 1.3638186356073213e-05, | |
| "loss": 0.0232, | |
| "step": 153000 | |
| }, | |
| { | |
| "epoch": 63.85, | |
| "learning_rate": 1.3617387687188022e-05, | |
| "loss": 0.0221, | |
| "step": 153500 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_cer": 0.2583, | |
| "eval_gen_len": 13.9485, | |
| "eval_loss": 0.3893982470035553, | |
| "eval_runtime": 270.1502, | |
| "eval_samples_per_second": 11.867, | |
| "eval_steps_per_second": 2.969, | |
| "step": 153856 | |
| }, | |
| { | |
| "epoch": 64.06, | |
| "learning_rate": 1.359658901830283e-05, | |
| "loss": 0.0268, | |
| "step": 154000 | |
| }, | |
| { | |
| "epoch": 64.27, | |
| "learning_rate": 1.3575790349417639e-05, | |
| "loss": 0.0159, | |
| "step": 154500 | |
| }, | |
| { | |
| "epoch": 64.48, | |
| "learning_rate": 1.3554991680532448e-05, | |
| "loss": 0.0191, | |
| "step": 155000 | |
| }, | |
| { | |
| "epoch": 64.68, | |
| "learning_rate": 1.3534193011647256e-05, | |
| "loss": 0.02, | |
| "step": 155500 | |
| }, | |
| { | |
| "epoch": 64.89, | |
| "learning_rate": 1.3513394342762065e-05, | |
| "loss": 0.0234, | |
| "step": 156000 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_cer": 0.2515, | |
| "eval_gen_len": 13.6329, | |
| "eval_loss": 0.39693862199783325, | |
| "eval_runtime": 262.2666, | |
| "eval_samples_per_second": 12.224, | |
| "eval_steps_per_second": 3.058, | |
| "step": 156260 | |
| }, | |
| { | |
| "epoch": 65.1, | |
| "learning_rate": 1.3492595673876873e-05, | |
| "loss": 0.0225, | |
| "step": 156500 | |
| }, | |
| { | |
| "epoch": 65.31, | |
| "learning_rate": 1.3471838602329453e-05, | |
| "loss": 0.0219, | |
| "step": 157000 | |
| }, | |
| { | |
| "epoch": 65.52, | |
| "learning_rate": 1.3451081530782031e-05, | |
| "loss": 0.0224, | |
| "step": 157500 | |
| }, | |
| { | |
| "epoch": 65.72, | |
| "learning_rate": 1.3430282861896839e-05, | |
| "loss": 0.0222, | |
| "step": 158000 | |
| }, | |
| { | |
| "epoch": 65.93, | |
| "learning_rate": 1.3409484193011648e-05, | |
| "loss": 0.0251, | |
| "step": 158500 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_cer": 0.2524, | |
| "eval_gen_len": 13.9046, | |
| "eval_loss": 0.3977407217025757, | |
| "eval_runtime": 268.0146, | |
| "eval_samples_per_second": 11.962, | |
| "eval_steps_per_second": 2.992, | |
| "step": 158664 | |
| }, | |
| { | |
| "epoch": 66.14, | |
| "learning_rate": 1.3388685524126458e-05, | |
| "loss": 0.0181, | |
| "step": 159000 | |
| }, | |
| { | |
| "epoch": 66.35, | |
| "learning_rate": 1.3367886855241265e-05, | |
| "loss": 0.0194, | |
| "step": 159500 | |
| }, | |
| { | |
| "epoch": 66.56, | |
| "learning_rate": 1.3347129783693843e-05, | |
| "loss": 0.0209, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 66.76, | |
| "learning_rate": 1.3326331114808653e-05, | |
| "loss": 0.0235, | |
| "step": 160500 | |
| }, | |
| { | |
| "epoch": 66.97, | |
| "learning_rate": 1.3305532445923462e-05, | |
| "loss": 0.0253, | |
| "step": 161000 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_cer": 0.2511, | |
| "eval_gen_len": 13.6563, | |
| "eval_loss": 0.40182340145111084, | |
| "eval_runtime": 252.4386, | |
| "eval_samples_per_second": 12.7, | |
| "eval_steps_per_second": 3.177, | |
| "step": 161068 | |
| }, | |
| { | |
| "epoch": 67.18, | |
| "learning_rate": 1.328473377703827e-05, | |
| "loss": 0.0218, | |
| "step": 161500 | |
| }, | |
| { | |
| "epoch": 67.39, | |
| "learning_rate": 1.326393510815308e-05, | |
| "loss": 0.0175, | |
| "step": 162000 | |
| }, | |
| { | |
| "epoch": 67.6, | |
| "learning_rate": 1.3243136439267887e-05, | |
| "loss": 0.0204, | |
| "step": 162500 | |
| }, | |
| { | |
| "epoch": 67.8, | |
| "learning_rate": 1.3222337770382696e-05, | |
| "loss": 0.0197, | |
| "step": 163000 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_cer": 0.2559, | |
| "eval_gen_len": 14.0237, | |
| "eval_loss": 0.3884351849555969, | |
| "eval_runtime": 272.5844, | |
| "eval_samples_per_second": 11.761, | |
| "eval_steps_per_second": 2.942, | |
| "step": 163472 | |
| }, | |
| { | |
| "epoch": 68.01, | |
| "learning_rate": 1.3201539101497504e-05, | |
| "loss": 0.0241, | |
| "step": 163500 | |
| }, | |
| { | |
| "epoch": 68.22, | |
| "learning_rate": 1.3180782029950084e-05, | |
| "loss": 0.0179, | |
| "step": 164000 | |
| }, | |
| { | |
| "epoch": 68.43, | |
| "learning_rate": 1.3159983361064892e-05, | |
| "loss": 0.0175, | |
| "step": 164500 | |
| }, | |
| { | |
| "epoch": 68.64, | |
| "learning_rate": 1.3139184692179701e-05, | |
| "loss": 0.0227, | |
| "step": 165000 | |
| }, | |
| { | |
| "epoch": 68.84, | |
| "learning_rate": 1.3118386023294509e-05, | |
| "loss": 0.0219, | |
| "step": 165500 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_cer": 0.2519, | |
| "eval_gen_len": 13.5009, | |
| "eval_loss": 0.402326375246048, | |
| "eval_runtime": 254.8714, | |
| "eval_samples_per_second": 12.579, | |
| "eval_steps_per_second": 3.147, | |
| "step": 165876 | |
| }, | |
| { | |
| "epoch": 69.05, | |
| "learning_rate": 1.309762895174709e-05, | |
| "loss": 0.0229, | |
| "step": 166000 | |
| }, | |
| { | |
| "epoch": 69.26, | |
| "learning_rate": 1.30768302828619e-05, | |
| "loss": 0.0183, | |
| "step": 166500 | |
| }, | |
| { | |
| "epoch": 69.47, | |
| "learning_rate": 1.3056073211314478e-05, | |
| "loss": 0.0195, | |
| "step": 167000 | |
| }, | |
| { | |
| "epoch": 69.68, | |
| "learning_rate": 1.3035274542429285e-05, | |
| "loss": 0.0204, | |
| "step": 167500 | |
| }, | |
| { | |
| "epoch": 69.88, | |
| "learning_rate": 1.3014475873544095e-05, | |
| "loss": 0.0207, | |
| "step": 168000 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_cer": 0.2559, | |
| "eval_gen_len": 13.4392, | |
| "eval_loss": 0.390462189912796, | |
| "eval_runtime": 247.3478, | |
| "eval_samples_per_second": 12.962, | |
| "eval_steps_per_second": 3.242, | |
| "step": 168280 | |
| }, | |
| { | |
| "epoch": 70.09, | |
| "learning_rate": 1.2993677204658904e-05, | |
| "loss": 0.0216, | |
| "step": 168500 | |
| }, | |
| { | |
| "epoch": 70.3, | |
| "learning_rate": 1.2972878535773712e-05, | |
| "loss": 0.0192, | |
| "step": 169000 | |
| }, | |
| { | |
| "epoch": 70.51, | |
| "learning_rate": 1.2952079866888521e-05, | |
| "loss": 0.0192, | |
| "step": 169500 | |
| }, | |
| { | |
| "epoch": 70.72, | |
| "learning_rate": 1.2931281198003329e-05, | |
| "loss": 0.0187, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 70.92, | |
| "learning_rate": 1.2910524126455907e-05, | |
| "loss": 0.0233, | |
| "step": 170500 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_cer": 0.2574, | |
| "eval_gen_len": 13.7012, | |
| "eval_loss": 0.4090117812156677, | |
| "eval_runtime": 257.6398, | |
| "eval_samples_per_second": 12.444, | |
| "eval_steps_per_second": 3.113, | |
| "step": 170684 | |
| }, | |
| { | |
| "epoch": 71.13, | |
| "learning_rate": 1.2889725457570716e-05, | |
| "loss": 0.019, | |
| "step": 171000 | |
| }, | |
| { | |
| "epoch": 71.34, | |
| "learning_rate": 1.2868926788685526e-05, | |
| "loss": 0.0159, | |
| "step": 171500 | |
| }, | |
| { | |
| "epoch": 71.55, | |
| "learning_rate": 1.2848128119800333e-05, | |
| "loss": 0.0213, | |
| "step": 172000 | |
| }, | |
| { | |
| "epoch": 71.76, | |
| "learning_rate": 1.2827329450915143e-05, | |
| "loss": 0.0177, | |
| "step": 172500 | |
| }, | |
| { | |
| "epoch": 71.96, | |
| "learning_rate": 1.280653078202995e-05, | |
| "loss": 0.024, | |
| "step": 173000 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_cer": 0.2472, | |
| "eval_gen_len": 13.6382, | |
| "eval_loss": 0.38612431287765503, | |
| "eval_runtime": 258.2841, | |
| "eval_samples_per_second": 12.413, | |
| "eval_steps_per_second": 3.105, | |
| "step": 173088 | |
| }, | |
| { | |
| "epoch": 72.17, | |
| "learning_rate": 1.278577371048253e-05, | |
| "loss": 0.0176, | |
| "step": 173500 | |
| }, | |
| { | |
| "epoch": 72.38, | |
| "learning_rate": 1.2764975041597338e-05, | |
| "loss": 0.017, | |
| "step": 174000 | |
| }, | |
| { | |
| "epoch": 72.59, | |
| "learning_rate": 1.2744176372712147e-05, | |
| "loss": 0.0167, | |
| "step": 174500 | |
| }, | |
| { | |
| "epoch": 72.8, | |
| "learning_rate": 1.2723377703826955e-05, | |
| "loss": 0.0219, | |
| "step": 175000 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_cer": 0.2647, | |
| "eval_gen_len": 14.1634, | |
| "eval_loss": 0.4713122546672821, | |
| "eval_runtime": 271.6072, | |
| "eval_samples_per_second": 11.804, | |
| "eval_steps_per_second": 2.953, | |
| "step": 175492 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "learning_rate": 1.2702579034941764e-05, | |
| "loss": 0.0198, | |
| "step": 175500 | |
| }, | |
| { | |
| "epoch": 73.21, | |
| "learning_rate": 1.2681780366056574e-05, | |
| "loss": 0.0147, | |
| "step": 176000 | |
| }, | |
| { | |
| "epoch": 73.42, | |
| "learning_rate": 1.2660981697171381e-05, | |
| "loss": 0.0181, | |
| "step": 176500 | |
| }, | |
| { | |
| "epoch": 73.63, | |
| "learning_rate": 1.264018302828619e-05, | |
| "loss": 0.02, | |
| "step": 177000 | |
| }, | |
| { | |
| "epoch": 73.84, | |
| "learning_rate": 1.2619384359400998e-05, | |
| "loss": 0.0189, | |
| "step": 177500 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_cer": 0.2536, | |
| "eval_gen_len": 13.9267, | |
| "eval_loss": 0.39804303646087646, | |
| "eval_runtime": 266.4263, | |
| "eval_samples_per_second": 12.033, | |
| "eval_steps_per_second": 3.01, | |
| "step": 177896 | |
| }, | |
| { | |
| "epoch": 74.04, | |
| "learning_rate": 1.2598627287853578e-05, | |
| "loss": 0.0215, | |
| "step": 178000 | |
| }, | |
| { | |
| "epoch": 74.25, | |
| "learning_rate": 1.2577828618968386e-05, | |
| "loss": 0.0159, | |
| "step": 178500 | |
| }, | |
| { | |
| "epoch": 74.46, | |
| "learning_rate": 1.2557029950083195e-05, | |
| "loss": 0.0188, | |
| "step": 179000 | |
| }, | |
| { | |
| "epoch": 74.67, | |
| "learning_rate": 1.2536231281198003e-05, | |
| "loss": 0.0169, | |
| "step": 179500 | |
| }, | |
| { | |
| "epoch": 74.88, | |
| "learning_rate": 1.2515474209650585e-05, | |
| "loss": 0.0162, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_cer": 0.2529, | |
| "eval_gen_len": 13.6974, | |
| "eval_loss": 0.39674171805381775, | |
| "eval_runtime": 262.8858, | |
| "eval_samples_per_second": 12.195, | |
| "eval_steps_per_second": 3.051, | |
| "step": 180300 | |
| }, | |
| { | |
| "epoch": 75.08, | |
| "learning_rate": 1.2494675540765392e-05, | |
| "loss": 0.0193, | |
| "step": 180500 | |
| }, | |
| { | |
| "epoch": 75.29, | |
| "learning_rate": 1.2473876871880202e-05, | |
| "loss": 0.0162, | |
| "step": 181000 | |
| }, | |
| { | |
| "epoch": 75.5, | |
| "learning_rate": 1.2453078202995011e-05, | |
| "loss": 0.0174, | |
| "step": 181500 | |
| }, | |
| { | |
| "epoch": 75.71, | |
| "learning_rate": 1.243232113144759e-05, | |
| "loss": 0.017, | |
| "step": 182000 | |
| }, | |
| { | |
| "epoch": 75.92, | |
| "learning_rate": 1.2411522462562397e-05, | |
| "loss": 0.0183, | |
| "step": 182500 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_cer": 0.2512, | |
| "eval_gen_len": 13.708, | |
| "eval_loss": 0.42503321170806885, | |
| "eval_runtime": 262.6259, | |
| "eval_samples_per_second": 12.207, | |
| "eval_steps_per_second": 3.054, | |
| "step": 182704 | |
| }, | |
| { | |
| "epoch": 76.12, | |
| "learning_rate": 1.2390765391014977e-05, | |
| "loss": 0.0172, | |
| "step": 183000 | |
| }, | |
| { | |
| "epoch": 76.33, | |
| "learning_rate": 1.2369966722129784e-05, | |
| "loss": 0.0164, | |
| "step": 183500 | |
| }, | |
| { | |
| "epoch": 76.54, | |
| "learning_rate": 1.2349168053244594e-05, | |
| "loss": 0.0179, | |
| "step": 184000 | |
| }, | |
| { | |
| "epoch": 76.75, | |
| "learning_rate": 1.2328369384359401e-05, | |
| "loss": 0.0181, | |
| "step": 184500 | |
| }, | |
| { | |
| "epoch": 76.96, | |
| "learning_rate": 1.230757071547421e-05, | |
| "loss": 0.0202, | |
| "step": 185000 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_cer": 0.2548, | |
| "eval_gen_len": 13.8085, | |
| "eval_loss": 0.4239508807659149, | |
| "eval_runtime": 252.2131, | |
| "eval_samples_per_second": 12.711, | |
| "eval_steps_per_second": 3.18, | |
| "step": 185108 | |
| }, | |
| { | |
| "epoch": 77.16, | |
| "learning_rate": 1.228677204658902e-05, | |
| "loss": 0.015, | |
| "step": 185500 | |
| }, | |
| { | |
| "epoch": 77.37, | |
| "learning_rate": 1.2265973377703828e-05, | |
| "loss": 0.0148, | |
| "step": 186000 | |
| }, | |
| { | |
| "epoch": 77.58, | |
| "learning_rate": 1.2245174708818637e-05, | |
| "loss": 0.0179, | |
| "step": 186500 | |
| }, | |
| { | |
| "epoch": 77.79, | |
| "learning_rate": 1.2224417637271215e-05, | |
| "loss": 0.019, | |
| "step": 187000 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "learning_rate": 1.2203660565723794e-05, | |
| "loss": 0.0186, | |
| "step": 187500 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_cer": 0.2522, | |
| "eval_gen_len": 13.806, | |
| "eval_loss": 0.4266161620616913, | |
| "eval_runtime": 261.9494, | |
| "eval_samples_per_second": 12.239, | |
| "eval_steps_per_second": 3.062, | |
| "step": 187512 | |
| }, | |
| { | |
| "epoch": 78.2, | |
| "learning_rate": 1.2182861896838603e-05, | |
| "loss": 0.0155, | |
| "step": 188000 | |
| }, | |
| { | |
| "epoch": 78.41, | |
| "learning_rate": 1.216206322795341e-05, | |
| "loss": 0.0167, | |
| "step": 188500 | |
| }, | |
| { | |
| "epoch": 78.62, | |
| "learning_rate": 1.214126455906822e-05, | |
| "loss": 0.0176, | |
| "step": 189000 | |
| }, | |
| { | |
| "epoch": 78.83, | |
| "learning_rate": 1.212046589018303e-05, | |
| "loss": 0.016, | |
| "step": 189500 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_cer": 0.2499, | |
| "eval_gen_len": 13.713, | |
| "eval_loss": 0.4089159667491913, | |
| "eval_runtime": 255.1636, | |
| "eval_samples_per_second": 12.564, | |
| "eval_steps_per_second": 3.143, | |
| "step": 189916 | |
| }, | |
| { | |
| "epoch": 79.03, | |
| "learning_rate": 1.2099667221297837e-05, | |
| "loss": 0.0165, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 79.24, | |
| "learning_rate": 1.2078868552412646e-05, | |
| "loss": 0.0136, | |
| "step": 190500 | |
| }, | |
| { | |
| "epoch": 79.45, | |
| "learning_rate": 1.2058111480865226e-05, | |
| "loss": 0.0154, | |
| "step": 191000 | |
| }, | |
| { | |
| "epoch": 79.66, | |
| "learning_rate": 1.2037312811980036e-05, | |
| "loss": 0.0168, | |
| "step": 191500 | |
| }, | |
| { | |
| "epoch": 79.87, | |
| "learning_rate": 1.2016514143094843e-05, | |
| "loss": 0.0188, | |
| "step": 192000 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_cer": 0.2501, | |
| "eval_gen_len": 13.7745, | |
| "eval_loss": 0.41349881887435913, | |
| "eval_runtime": 263.2542, | |
| "eval_samples_per_second": 12.178, | |
| "eval_steps_per_second": 3.046, | |
| "step": 192320 | |
| }, | |
| { | |
| "epoch": 80.07, | |
| "learning_rate": 1.1995715474209653e-05, | |
| "loss": 0.0169, | |
| "step": 192500 | |
| }, | |
| { | |
| "epoch": 80.28, | |
| "learning_rate": 1.1974916805324459e-05, | |
| "loss": 0.0142, | |
| "step": 193000 | |
| }, | |
| { | |
| "epoch": 80.49, | |
| "learning_rate": 1.1954118136439268e-05, | |
| "loss": 0.0186, | |
| "step": 193500 | |
| }, | |
| { | |
| "epoch": 80.7, | |
| "learning_rate": 1.1933319467554076e-05, | |
| "loss": 0.0173, | |
| "step": 194000 | |
| }, | |
| { | |
| "epoch": 80.91, | |
| "learning_rate": 1.1912520798668885e-05, | |
| "loss": 0.016, | |
| "step": 194500 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_cer": 0.2477, | |
| "eval_gen_len": 13.6622, | |
| "eval_loss": 0.3864258825778961, | |
| "eval_runtime": 259.6006, | |
| "eval_samples_per_second": 12.35, | |
| "eval_steps_per_second": 3.089, | |
| "step": 194724 | |
| }, | |
| { | |
| "epoch": 81.11, | |
| "learning_rate": 1.1891763727121467e-05, | |
| "loss": 0.0166, | |
| "step": 195000 | |
| }, | |
| { | |
| "epoch": 81.32, | |
| "learning_rate": 1.1870965058236274e-05, | |
| "loss": 0.0141, | |
| "step": 195500 | |
| }, | |
| { | |
| "epoch": 81.53, | |
| "learning_rate": 1.1850207986688853e-05, | |
| "loss": 0.0151, | |
| "step": 196000 | |
| }, | |
| { | |
| "epoch": 81.74, | |
| "learning_rate": 1.1829409317803662e-05, | |
| "loss": 0.0176, | |
| "step": 196500 | |
| }, | |
| { | |
| "epoch": 81.95, | |
| "learning_rate": 1.180861064891847e-05, | |
| "loss": 0.0156, | |
| "step": 197000 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_cer": 0.2551, | |
| "eval_gen_len": 13.9261, | |
| "eval_loss": 0.4278740882873535, | |
| "eval_runtime": 261.2997, | |
| "eval_samples_per_second": 12.269, | |
| "eval_steps_per_second": 3.069, | |
| "step": 197128 | |
| }, | |
| { | |
| "epoch": 82.15, | |
| "learning_rate": 1.1787811980033279e-05, | |
| "loss": 0.0173, | |
| "step": 197500 | |
| }, | |
| { | |
| "epoch": 82.36, | |
| "learning_rate": 1.1767013311148088e-05, | |
| "loss": 0.0153, | |
| "step": 198000 | |
| }, | |
| { | |
| "epoch": 82.57, | |
| "learning_rate": 1.1746214642262896e-05, | |
| "loss": 0.016, | |
| "step": 198500 | |
| }, | |
| { | |
| "epoch": 82.78, | |
| "learning_rate": 1.1725415973377705e-05, | |
| "loss": 0.0136, | |
| "step": 199000 | |
| }, | |
| { | |
| "epoch": 82.99, | |
| "learning_rate": 1.1704617304492513e-05, | |
| "loss": 0.018, | |
| "step": 199500 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_cer": 0.25, | |
| "eval_gen_len": 13.704, | |
| "eval_loss": 0.4216358959674835, | |
| "eval_runtime": 257.2522, | |
| "eval_samples_per_second": 12.462, | |
| "eval_steps_per_second": 3.118, | |
| "step": 199532 | |
| }, | |
| { | |
| "epoch": 83.19, | |
| "learning_rate": 1.1683818635607322e-05, | |
| "loss": 0.0125, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 83.4, | |
| "learning_rate": 1.16630615640599e-05, | |
| "loss": 0.0158, | |
| "step": 200500 | |
| }, | |
| { | |
| "epoch": 83.61, | |
| "learning_rate": 1.1642304492512479e-05, | |
| "loss": 0.0153, | |
| "step": 201000 | |
| }, | |
| { | |
| "epoch": 83.82, | |
| "learning_rate": 1.1621505823627288e-05, | |
| "loss": 0.0159, | |
| "step": 201500 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_cer": 0.2502, | |
| "eval_gen_len": 13.7121, | |
| "eval_loss": 0.42220476269721985, | |
| "eval_runtime": 260.9684, | |
| "eval_samples_per_second": 12.285, | |
| "eval_steps_per_second": 3.073, | |
| "step": 201936 | |
| }, | |
| { | |
| "epoch": 84.03, | |
| "learning_rate": 1.1600707154742097e-05, | |
| "loss": 0.0159, | |
| "step": 202000 | |
| }, | |
| { | |
| "epoch": 84.23, | |
| "learning_rate": 1.1579908485856905e-05, | |
| "loss": 0.0157, | |
| "step": 202500 | |
| }, | |
| { | |
| "epoch": 84.44, | |
| "learning_rate": 1.1559109816971715e-05, | |
| "loss": 0.0149, | |
| "step": 203000 | |
| }, | |
| { | |
| "epoch": 84.65, | |
| "learning_rate": 1.1538311148086522e-05, | |
| "loss": 0.0148, | |
| "step": 203500 | |
| }, | |
| { | |
| "epoch": 84.86, | |
| "learning_rate": 1.1517512479201332e-05, | |
| "loss": 0.0165, | |
| "step": 204000 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_cer": 0.2482, | |
| "eval_gen_len": 13.7233, | |
| "eval_loss": 0.4069821238517761, | |
| "eval_runtime": 256.2579, | |
| "eval_samples_per_second": 12.511, | |
| "eval_steps_per_second": 3.13, | |
| "step": 204340 | |
| }, | |
| { | |
| "epoch": 85.07, | |
| "learning_rate": 1.149675540765391e-05, | |
| "loss": 0.0162, | |
| "step": 204500 | |
| }, | |
| { | |
| "epoch": 85.27, | |
| "learning_rate": 1.1475956738768719e-05, | |
| "loss": 0.0128, | |
| "step": 205000 | |
| }, | |
| { | |
| "epoch": 85.48, | |
| "learning_rate": 1.1455158069883527e-05, | |
| "loss": 0.0149, | |
| "step": 205500 | |
| }, | |
| { | |
| "epoch": 85.69, | |
| "learning_rate": 1.1434359400998336e-05, | |
| "loss": 0.0182, | |
| "step": 206000 | |
| }, | |
| { | |
| "epoch": 85.9, | |
| "learning_rate": 1.1413560732113146e-05, | |
| "loss": 0.0149, | |
| "step": 206500 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_cer": 0.2493, | |
| "eval_gen_len": 13.7623, | |
| "eval_loss": 0.4060095250606537, | |
| "eval_runtime": 264.6587, | |
| "eval_samples_per_second": 12.114, | |
| "eval_steps_per_second": 3.03, | |
| "step": 206744 | |
| }, | |
| { | |
| "epoch": 86.11, | |
| "learning_rate": 1.1392762063227953e-05, | |
| "loss": 0.0131, | |
| "step": 207000 | |
| }, | |
| { | |
| "epoch": 86.31, | |
| "learning_rate": 1.1371963394342763e-05, | |
| "loss": 0.0145, | |
| "step": 207500 | |
| }, | |
| { | |
| "epoch": 86.52, | |
| "learning_rate": 1.135116472545757e-05, | |
| "loss": 0.0133, | |
| "step": 208000 | |
| }, | |
| { | |
| "epoch": 86.73, | |
| "learning_rate": 1.1330407653910152e-05, | |
| "loss": 0.0141, | |
| "step": 208500 | |
| }, | |
| { | |
| "epoch": 86.94, | |
| "learning_rate": 1.130960898502496e-05, | |
| "loss": 0.014, | |
| "step": 209000 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_cer": 0.2461, | |
| "eval_gen_len": 13.6067, | |
| "eval_loss": 0.42620450258255005, | |
| "eval_runtime": 258.5687, | |
| "eval_samples_per_second": 12.399, | |
| "eval_steps_per_second": 3.102, | |
| "step": 209148 | |
| }, | |
| { | |
| "epoch": 87.15, | |
| "learning_rate": 1.1288810316139769e-05, | |
| "loss": 0.0131, | |
| "step": 209500 | |
| }, | |
| { | |
| "epoch": 87.35, | |
| "learning_rate": 1.1268011647254578e-05, | |
| "loss": 0.0121, | |
| "step": 210000 | |
| }, | |
| { | |
| "epoch": 87.56, | |
| "learning_rate": 1.1247212978369386e-05, | |
| "loss": 0.015, | |
| "step": 210500 | |
| }, | |
| { | |
| "epoch": 87.77, | |
| "learning_rate": 1.1226414309484195e-05, | |
| "loss": 0.0137, | |
| "step": 211000 | |
| }, | |
| { | |
| "epoch": 87.98, | |
| "learning_rate": 1.1205615640599003e-05, | |
| "loss": 0.0161, | |
| "step": 211500 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_cer": 0.249, | |
| "eval_gen_len": 13.758, | |
| "eval_loss": 0.4252397418022156, | |
| "eval_runtime": 265.7747, | |
| "eval_samples_per_second": 12.063, | |
| "eval_steps_per_second": 3.018, | |
| "step": 211552 | |
| }, | |
| { | |
| "epoch": 88.19, | |
| "learning_rate": 1.1184816971713812e-05, | |
| "loss": 0.0126, | |
| "step": 212000 | |
| }, | |
| { | |
| "epoch": 88.39, | |
| "learning_rate": 1.116405990016639e-05, | |
| "loss": 0.0136, | |
| "step": 212500 | |
| }, | |
| { | |
| "epoch": 88.6, | |
| "learning_rate": 1.1143302828618969e-05, | |
| "loss": 0.0141, | |
| "step": 213000 | |
| }, | |
| { | |
| "epoch": 88.81, | |
| "learning_rate": 1.1122504159733778e-05, | |
| "loss": 0.0142, | |
| "step": 213500 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_cer": 0.2511, | |
| "eval_gen_len": 13.8013, | |
| "eval_loss": 0.43667590618133545, | |
| "eval_runtime": 258.922, | |
| "eval_samples_per_second": 12.382, | |
| "eval_steps_per_second": 3.097, | |
| "step": 213956 | |
| }, | |
| { | |
| "epoch": 89.02, | |
| "learning_rate": 1.1101705490848587e-05, | |
| "loss": 0.0155, | |
| "step": 214000 | |
| }, | |
| { | |
| "epoch": 89.23, | |
| "learning_rate": 1.1080906821963395e-05, | |
| "loss": 0.0149, | |
| "step": 214500 | |
| }, | |
| { | |
| "epoch": 89.43, | |
| "learning_rate": 1.1060108153078204e-05, | |
| "loss": 0.0138, | |
| "step": 215000 | |
| }, | |
| { | |
| "epoch": 89.64, | |
| "learning_rate": 1.1039309484193012e-05, | |
| "loss": 0.0131, | |
| "step": 215500 | |
| }, | |
| { | |
| "epoch": 89.85, | |
| "learning_rate": 1.1018510815307822e-05, | |
| "loss": 0.0146, | |
| "step": 216000 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_cer": 0.2483, | |
| "eval_gen_len": 13.6778, | |
| "eval_loss": 0.41625672578811646, | |
| "eval_runtime": 260.915, | |
| "eval_samples_per_second": 12.288, | |
| "eval_steps_per_second": 3.074, | |
| "step": 216360 | |
| }, | |
| { | |
| "epoch": 90.06, | |
| "learning_rate": 1.099771214642263e-05, | |
| "loss": 0.0131, | |
| "step": 216500 | |
| }, | |
| { | |
| "epoch": 90.27, | |
| "learning_rate": 1.0976955074875209e-05, | |
| "loss": 0.0128, | |
| "step": 217000 | |
| }, | |
| { | |
| "epoch": 90.47, | |
| "learning_rate": 1.0956156405990017e-05, | |
| "loss": 0.0139, | |
| "step": 217500 | |
| }, | |
| { | |
| "epoch": 90.68, | |
| "learning_rate": 1.0935357737104826e-05, | |
| "loss": 0.0146, | |
| "step": 218000 | |
| }, | |
| { | |
| "epoch": 90.89, | |
| "learning_rate": 1.0914559068219634e-05, | |
| "loss": 0.0127, | |
| "step": 218500 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_cer": 0.2466, | |
| "eval_gen_len": 13.6344, | |
| "eval_loss": 0.42400336265563965, | |
| "eval_runtime": 259.4085, | |
| "eval_samples_per_second": 12.359, | |
| "eval_steps_per_second": 3.092, | |
| "step": 218764 | |
| }, | |
| { | |
| "epoch": 91.1, | |
| "learning_rate": 1.0893843594009986e-05, | |
| "loss": 0.0131, | |
| "step": 219000 | |
| }, | |
| { | |
| "epoch": 91.31, | |
| "learning_rate": 1.0873044925124794e-05, | |
| "loss": 0.0139, | |
| "step": 219500 | |
| }, | |
| { | |
| "epoch": 91.51, | |
| "learning_rate": 1.0852246256239603e-05, | |
| "loss": 0.0144, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 91.72, | |
| "learning_rate": 1.083144758735441e-05, | |
| "loss": 0.016, | |
| "step": 220500 | |
| }, | |
| { | |
| "epoch": 91.93, | |
| "learning_rate": 1.081064891846922e-05, | |
| "loss": 0.0147, | |
| "step": 221000 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_cer": 0.2457, | |
| "eval_gen_len": 13.5948, | |
| "eval_loss": 0.4094228148460388, | |
| "eval_runtime": 257.4623, | |
| "eval_samples_per_second": 12.452, | |
| "eval_steps_per_second": 3.115, | |
| "step": 221168 | |
| }, | |
| { | |
| "epoch": 92.14, | |
| "learning_rate": 1.078985024958403e-05, | |
| "loss": 0.0136, | |
| "step": 221500 | |
| }, | |
| { | |
| "epoch": 92.35, | |
| "learning_rate": 1.0769051580698837e-05, | |
| "loss": 0.0123, | |
| "step": 222000 | |
| }, | |
| { | |
| "epoch": 92.55, | |
| "learning_rate": 1.0748252911813646e-05, | |
| "loss": 0.0113, | |
| "step": 222500 | |
| }, | |
| { | |
| "epoch": 92.76, | |
| "learning_rate": 1.0727454242928454e-05, | |
| "loss": 0.0153, | |
| "step": 223000 | |
| }, | |
| { | |
| "epoch": 92.97, | |
| "learning_rate": 1.0706697171381034e-05, | |
| "loss": 0.0153, | |
| "step": 223500 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_cer": 0.2414, | |
| "eval_gen_len": 13.5168, | |
| "eval_loss": 0.419572651386261, | |
| "eval_runtime": 254.173, | |
| "eval_samples_per_second": 12.613, | |
| "eval_steps_per_second": 3.155, | |
| "step": 223572 | |
| }, | |
| { | |
| "epoch": 93.18, | |
| "learning_rate": 1.0685898502495842e-05, | |
| "loss": 0.0131, | |
| "step": 224000 | |
| }, | |
| { | |
| "epoch": 93.39, | |
| "learning_rate": 1.0665099833610651e-05, | |
| "loss": 0.0138, | |
| "step": 224500 | |
| }, | |
| { | |
| "epoch": 93.59, | |
| "learning_rate": 1.0644301164725459e-05, | |
| "loss": 0.0125, | |
| "step": 225000 | |
| }, | |
| { | |
| "epoch": 93.8, | |
| "learning_rate": 1.0623544093178037e-05, | |
| "loss": 0.0158, | |
| "step": 225500 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_cer": 0.2491, | |
| "eval_gen_len": 13.8378, | |
| "eval_loss": 0.4395461678504944, | |
| "eval_runtime": 257.6708, | |
| "eval_samples_per_second": 12.442, | |
| "eval_steps_per_second": 3.112, | |
| "step": 225976 | |
| }, | |
| { | |
| "epoch": 94.01, | |
| "learning_rate": 1.0602745424292846e-05, | |
| "loss": 0.0141, | |
| "step": 226000 | |
| }, | |
| { | |
| "epoch": 94.22, | |
| "learning_rate": 1.0581946755407656e-05, | |
| "loss": 0.0147, | |
| "step": 226500 | |
| }, | |
| { | |
| "epoch": 94.43, | |
| "learning_rate": 1.0561148086522463e-05, | |
| "loss": 0.0119, | |
| "step": 227000 | |
| }, | |
| { | |
| "epoch": 94.63, | |
| "learning_rate": 1.0540391014975041e-05, | |
| "loss": 0.0139, | |
| "step": 227500 | |
| }, | |
| { | |
| "epoch": 94.84, | |
| "learning_rate": 1.0519633943427621e-05, | |
| "loss": 0.0138, | |
| "step": 228000 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_cer": 0.2518, | |
| "eval_gen_len": 13.7695, | |
| "eval_loss": 0.43901219964027405, | |
| "eval_runtime": 262.2107, | |
| "eval_samples_per_second": 12.227, | |
| "eval_steps_per_second": 3.059, | |
| "step": 228380 | |
| }, | |
| { | |
| "epoch": 95.05, | |
| "learning_rate": 1.0498835274542429e-05, | |
| "loss": 0.0121, | |
| "step": 228500 | |
| }, | |
| { | |
| "epoch": 95.26, | |
| "learning_rate": 1.0478036605657238e-05, | |
| "loss": 0.0106, | |
| "step": 229000 | |
| }, | |
| { | |
| "epoch": 95.47, | |
| "learning_rate": 1.0457237936772046e-05, | |
| "loss": 0.0126, | |
| "step": 229500 | |
| }, | |
| { | |
| "epoch": 95.67, | |
| "learning_rate": 1.0436439267886855e-05, | |
| "loss": 0.0156, | |
| "step": 230000 | |
| }, | |
| { | |
| "epoch": 95.88, | |
| "learning_rate": 1.0415640599001665e-05, | |
| "loss": 0.0123, | |
| "step": 230500 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_cer": 0.2475, | |
| "eval_gen_len": 13.69, | |
| "eval_loss": 0.4405384361743927, | |
| "eval_runtime": 260.5068, | |
| "eval_samples_per_second": 12.307, | |
| "eval_steps_per_second": 3.079, | |
| "step": 230784 | |
| }, | |
| { | |
| "epoch": 96.09, | |
| "learning_rate": 1.0394841930116472e-05, | |
| "loss": 0.011, | |
| "step": 231000 | |
| }, | |
| { | |
| "epoch": 96.3, | |
| "learning_rate": 1.0374084858569054e-05, | |
| "loss": 0.0132, | |
| "step": 231500 | |
| }, | |
| { | |
| "epoch": 96.51, | |
| "learning_rate": 1.0353286189683862e-05, | |
| "loss": 0.0135, | |
| "step": 232000 | |
| }, | |
| { | |
| "epoch": 96.71, | |
| "learning_rate": 1.0332487520798671e-05, | |
| "loss": 0.0143, | |
| "step": 232500 | |
| }, | |
| { | |
| "epoch": 96.92, | |
| "learning_rate": 1.0311688851913479e-05, | |
| "loss": 0.0136, | |
| "step": 233000 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_cer": 0.2492, | |
| "eval_gen_len": 13.791, | |
| "eval_loss": 0.4154476225376129, | |
| "eval_runtime": 266.6041, | |
| "eval_samples_per_second": 12.025, | |
| "eval_steps_per_second": 3.008, | |
| "step": 233188 | |
| }, | |
| { | |
| "epoch": 97.13, | |
| "learning_rate": 1.0290890183028288e-05, | |
| "loss": 0.0117, | |
| "step": 233500 | |
| }, | |
| { | |
| "epoch": 97.34, | |
| "learning_rate": 1.0270091514143094e-05, | |
| "loss": 0.0116, | |
| "step": 234000 | |
| }, | |
| { | |
| "epoch": 97.55, | |
| "learning_rate": 1.0249292845257903e-05, | |
| "loss": 0.0117, | |
| "step": 234500 | |
| }, | |
| { | |
| "epoch": 97.75, | |
| "learning_rate": 1.0228494176372713e-05, | |
| "loss": 0.0158, | |
| "step": 235000 | |
| }, | |
| { | |
| "epoch": 97.96, | |
| "learning_rate": 1.0207737104825293e-05, | |
| "loss": 0.012, | |
| "step": 235500 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_cer": 0.2481, | |
| "eval_gen_len": 13.8702, | |
| "eval_loss": 0.43725699186325073, | |
| "eval_runtime": 257.4309, | |
| "eval_samples_per_second": 12.454, | |
| "eval_steps_per_second": 3.115, | |
| "step": 235592 | |
| }, | |
| { | |
| "epoch": 98.17, | |
| "learning_rate": 1.0186938435940102e-05, | |
| "loss": 0.0098, | |
| "step": 236000 | |
| }, | |
| { | |
| "epoch": 98.38, | |
| "learning_rate": 1.016618136439268e-05, | |
| "loss": 0.0118, | |
| "step": 236500 | |
| }, | |
| { | |
| "epoch": 98.59, | |
| "learning_rate": 1.0145382695507488e-05, | |
| "loss": 0.0128, | |
| "step": 237000 | |
| }, | |
| { | |
| "epoch": 98.79, | |
| "learning_rate": 1.0124584026622297e-05, | |
| "loss": 0.0122, | |
| "step": 237500 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_cer": 0.2504, | |
| "eval_gen_len": 13.8051, | |
| "eval_loss": 0.44157010316848755, | |
| "eval_runtime": 284.8185, | |
| "eval_samples_per_second": 11.256, | |
| "eval_steps_per_second": 2.816, | |
| "step": 237996 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "learning_rate": 1.0103785357737107e-05, | |
| "loss": 0.0136, | |
| "step": 238000 | |
| }, | |
| { | |
| "epoch": 99.21, | |
| "learning_rate": 1.0082986688851914e-05, | |
| "loss": 0.0115, | |
| "step": 238500 | |
| }, | |
| { | |
| "epoch": 99.42, | |
| "learning_rate": 1.0062188019966724e-05, | |
| "loss": 0.0109, | |
| "step": 239000 | |
| }, | |
| { | |
| "epoch": 99.63, | |
| "learning_rate": 1.0041389351081531e-05, | |
| "loss": 0.0135, | |
| "step": 239500 | |
| }, | |
| { | |
| "epoch": 99.83, | |
| "learning_rate": 1.002059068219634e-05, | |
| "loss": 0.0146, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_cer": 0.2489, | |
| "eval_gen_len": 13.6868, | |
| "eval_loss": 0.42966365814208984, | |
| "eval_runtime": 282.168, | |
| "eval_samples_per_second": 11.362, | |
| "eval_steps_per_second": 2.842, | |
| "step": 240400 | |
| }, | |
| { | |
| "epoch": 100.04, | |
| "learning_rate": 9.99979201331115e-06, | |
| "loss": 0.0136, | |
| "step": 240500 | |
| }, | |
| { | |
| "epoch": 100.25, | |
| "learning_rate": 9.979034941763728e-06, | |
| "loss": 0.0132, | |
| "step": 241000 | |
| }, | |
| { | |
| "epoch": 100.46, | |
| "learning_rate": 9.958236272878536e-06, | |
| "loss": 0.0112, | |
| "step": 241500 | |
| }, | |
| { | |
| "epoch": 100.67, | |
| "learning_rate": 9.937437603993345e-06, | |
| "loss": 0.0139, | |
| "step": 242000 | |
| }, | |
| { | |
| "epoch": 100.87, | |
| "learning_rate": 9.916638935108155e-06, | |
| "loss": 0.0135, | |
| "step": 242500 | |
| }, | |
| { | |
| "epoch": 101.0, | |
| "eval_cer": 0.2428, | |
| "eval_gen_len": 13.5689, | |
| "eval_loss": 0.4310346245765686, | |
| "eval_runtime": 280.207, | |
| "eval_samples_per_second": 11.442, | |
| "eval_steps_per_second": 2.862, | |
| "step": 242804 | |
| }, | |
| { | |
| "epoch": 101.08, | |
| "learning_rate": 9.895881863560733e-06, | |
| "loss": 0.0118, | |
| "step": 243000 | |
| }, | |
| { | |
| "epoch": 101.29, | |
| "learning_rate": 9.875083194675542e-06, | |
| "loss": 0.0111, | |
| "step": 243500 | |
| }, | |
| { | |
| "epoch": 101.5, | |
| "learning_rate": 9.854284525790352e-06, | |
| "loss": 0.0115, | |
| "step": 244000 | |
| }, | |
| { | |
| "epoch": 101.71, | |
| "learning_rate": 9.83352745424293e-06, | |
| "loss": 0.0124, | |
| "step": 244500 | |
| }, | |
| { | |
| "epoch": 101.91, | |
| "learning_rate": 9.812728785357737e-06, | |
| "loss": 0.0136, | |
| "step": 245000 | |
| }, | |
| { | |
| "epoch": 102.0, | |
| "eval_cer": 0.246, | |
| "eval_gen_len": 13.4972, | |
| "eval_loss": 0.422376424074173, | |
| "eval_runtime": 269.57, | |
| "eval_samples_per_second": 11.893, | |
| "eval_steps_per_second": 2.975, | |
| "step": 245208 | |
| }, | |
| { | |
| "epoch": 102.12, | |
| "learning_rate": 9.791930116472547e-06, | |
| "loss": 0.0124, | |
| "step": 245500 | |
| }, | |
| { | |
| "epoch": 102.33, | |
| "learning_rate": 9.771131447587356e-06, | |
| "loss": 0.0101, | |
| "step": 246000 | |
| }, | |
| { | |
| "epoch": 102.54, | |
| "learning_rate": 9.750332778702164e-06, | |
| "loss": 0.0109, | |
| "step": 246500 | |
| }, | |
| { | |
| "epoch": 102.75, | |
| "learning_rate": 9.729534109816973e-06, | |
| "loss": 0.0108, | |
| "step": 247000 | |
| }, | |
| { | |
| "epoch": 102.95, | |
| "learning_rate": 9.70873544093178e-06, | |
| "loss": 0.014, | |
| "step": 247500 | |
| }, | |
| { | |
| "epoch": 103.0, | |
| "eval_cer": 0.2502, | |
| "eval_gen_len": 13.8394, | |
| "eval_loss": 0.45676541328430176, | |
| "eval_runtime": 270.1526, | |
| "eval_samples_per_second": 11.867, | |
| "eval_steps_per_second": 2.969, | |
| "step": 247612 | |
| }, | |
| { | |
| "epoch": 103.16, | |
| "learning_rate": 9.68793677204659e-06, | |
| "loss": 0.0106, | |
| "step": 248000 | |
| }, | |
| { | |
| "epoch": 103.37, | |
| "learning_rate": 9.667179700499168e-06, | |
| "loss": 0.0117, | |
| "step": 248500 | |
| }, | |
| { | |
| "epoch": 103.58, | |
| "learning_rate": 9.646381031613978e-06, | |
| "loss": 0.0114, | |
| "step": 249000 | |
| }, | |
| { | |
| "epoch": 103.79, | |
| "learning_rate": 9.625582362728785e-06, | |
| "loss": 0.0135, | |
| "step": 249500 | |
| }, | |
| { | |
| "epoch": 103.99, | |
| "learning_rate": 9.604783693843595e-06, | |
| "loss": 0.0125, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 104.0, | |
| "eval_cer": 0.2465, | |
| "eval_gen_len": 13.4345, | |
| "eval_loss": 0.39919513463974, | |
| "eval_runtime": 270.6771, | |
| "eval_samples_per_second": 11.844, | |
| "eval_steps_per_second": 2.963, | |
| "step": 250016 | |
| }, | |
| { | |
| "epoch": 104.2, | |
| "learning_rate": 9.583985024958402e-06, | |
| "loss": 0.01, | |
| "step": 250500 | |
| }, | |
| { | |
| "epoch": 104.41, | |
| "learning_rate": 9.563227953410982e-06, | |
| "loss": 0.0126, | |
| "step": 251000 | |
| }, | |
| { | |
| "epoch": 104.62, | |
| "learning_rate": 9.542429284525792e-06, | |
| "loss": 0.0109, | |
| "step": 251500 | |
| }, | |
| { | |
| "epoch": 104.83, | |
| "learning_rate": 9.5216306156406e-06, | |
| "loss": 0.0135, | |
| "step": 252000 | |
| }, | |
| { | |
| "epoch": 105.0, | |
| "eval_cer": 0.2472, | |
| "eval_gen_len": 13.7277, | |
| "eval_loss": 0.44164207577705383, | |
| "eval_runtime": 274.8419, | |
| "eval_samples_per_second": 11.665, | |
| "eval_steps_per_second": 2.918, | |
| "step": 252420 | |
| }, | |
| { | |
| "epoch": 105.03, | |
| "learning_rate": 9.500831946755409e-06, | |
| "loss": 0.0119, | |
| "step": 252500 | |
| }, | |
| { | |
| "epoch": 105.24, | |
| "learning_rate": 9.480033277870218e-06, | |
| "loss": 0.0095, | |
| "step": 253000 | |
| }, | |
| { | |
| "epoch": 105.45, | |
| "learning_rate": 9.459276206322796e-06, | |
| "loss": 0.0115, | |
| "step": 253500 | |
| }, | |
| { | |
| "epoch": 105.66, | |
| "learning_rate": 9.438477537437604e-06, | |
| "loss": 0.0135, | |
| "step": 254000 | |
| }, | |
| { | |
| "epoch": 105.87, | |
| "learning_rate": 9.417678868552413e-06, | |
| "loss": 0.012, | |
| "step": 254500 | |
| }, | |
| { | |
| "epoch": 106.0, | |
| "eval_cer": 0.2416, | |
| "eval_gen_len": 13.4994, | |
| "eval_loss": 0.41192150115966797, | |
| "eval_runtime": 275.6326, | |
| "eval_samples_per_second": 11.631, | |
| "eval_steps_per_second": 2.91, | |
| "step": 254824 | |
| }, | |
| { | |
| "epoch": 106.07, | |
| "learning_rate": 9.396880199667223e-06, | |
| "loss": 0.0114, | |
| "step": 255000 | |
| }, | |
| { | |
| "epoch": 106.28, | |
| "learning_rate": 9.376123128119801e-06, | |
| "loss": 0.0116, | |
| "step": 255500 | |
| }, | |
| { | |
| "epoch": 106.49, | |
| "learning_rate": 9.355324459234609e-06, | |
| "loss": 0.0093, | |
| "step": 256000 | |
| }, | |
| { | |
| "epoch": 106.7, | |
| "learning_rate": 9.334525790349418e-06, | |
| "loss": 0.0128, | |
| "step": 256500 | |
| }, | |
| { | |
| "epoch": 106.91, | |
| "learning_rate": 9.313768718801998e-06, | |
| "loss": 0.0133, | |
| "step": 257000 | |
| }, | |
| { | |
| "epoch": 107.0, | |
| "eval_cer": 0.2476, | |
| "eval_gen_len": 13.6494, | |
| "eval_loss": 0.42318016290664673, | |
| "eval_runtime": 280.8295, | |
| "eval_samples_per_second": 11.416, | |
| "eval_steps_per_second": 2.856, | |
| "step": 257228 | |
| }, | |
| { | |
| "epoch": 107.11, | |
| "learning_rate": 9.292970049916805e-06, | |
| "loss": 0.0097, | |
| "step": 257500 | |
| }, | |
| { | |
| "epoch": 107.32, | |
| "learning_rate": 9.272171381031615e-06, | |
| "loss": 0.012, | |
| "step": 258000 | |
| }, | |
| { | |
| "epoch": 107.53, | |
| "learning_rate": 9.251372712146424e-06, | |
| "loss": 0.0093, | |
| "step": 258500 | |
| }, | |
| { | |
| "epoch": 107.74, | |
| "learning_rate": 9.230574043261232e-06, | |
| "loss": 0.0103, | |
| "step": 259000 | |
| }, | |
| { | |
| "epoch": 107.95, | |
| "learning_rate": 9.209775374376041e-06, | |
| "loss": 0.0103, | |
| "step": 259500 | |
| }, | |
| { | |
| "epoch": 108.0, | |
| "eval_cer": 0.2434, | |
| "eval_gen_len": 13.4925, | |
| "eval_loss": 0.425822377204895, | |
| "eval_runtime": 284.9328, | |
| "eval_samples_per_second": 11.252, | |
| "eval_steps_per_second": 2.815, | |
| "step": 259632 | |
| }, | |
| { | |
| "epoch": 108.15, | |
| "learning_rate": 9.188976705490849e-06, | |
| "loss": 0.0108, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 108.36, | |
| "learning_rate": 9.168178036605658e-06, | |
| "loss": 0.0109, | |
| "step": 260500 | |
| }, | |
| { | |
| "epoch": 108.57, | |
| "learning_rate": 9.147420965058236e-06, | |
| "loss": 0.0125, | |
| "step": 261000 | |
| }, | |
| { | |
| "epoch": 108.78, | |
| "learning_rate": 9.126622296173046e-06, | |
| "loss": 0.0112, | |
| "step": 261500 | |
| }, | |
| { | |
| "epoch": 108.99, | |
| "learning_rate": 9.105823627287854e-06, | |
| "loss": 0.0116, | |
| "step": 262000 | |
| }, | |
| { | |
| "epoch": 109.0, | |
| "eval_cer": 0.2462, | |
| "eval_gen_len": 13.7115, | |
| "eval_loss": 0.43990305066108704, | |
| "eval_runtime": 285.2346, | |
| "eval_samples_per_second": 11.24, | |
| "eval_steps_per_second": 2.812, | |
| "step": 262036 | |
| }, | |
| { | |
| "epoch": 109.19, | |
| "learning_rate": 9.085066555740433e-06, | |
| "loss": 0.0088, | |
| "step": 262500 | |
| }, | |
| { | |
| "epoch": 109.4, | |
| "learning_rate": 9.064267886855243e-06, | |
| "loss": 0.0098, | |
| "step": 263000 | |
| }, | |
| { | |
| "epoch": 109.61, | |
| "learning_rate": 9.04346921797005e-06, | |
| "loss": 0.0106, | |
| "step": 263500 | |
| }, | |
| { | |
| "epoch": 109.82, | |
| "learning_rate": 9.02267054908486e-06, | |
| "loss": 0.0123, | |
| "step": 264000 | |
| }, | |
| { | |
| "epoch": 110.0, | |
| "eval_cer": 0.2462, | |
| "eval_gen_len": 13.6023, | |
| "eval_loss": 0.41700801253318787, | |
| "eval_runtime": 287.8449, | |
| "eval_samples_per_second": 11.138, | |
| "eval_steps_per_second": 2.786, | |
| "step": 264440 | |
| }, | |
| { | |
| "epoch": 110.02, | |
| "learning_rate": 9.00187188019967e-06, | |
| "loss": 0.0125, | |
| "step": 264500 | |
| }, | |
| { | |
| "epoch": 110.23, | |
| "learning_rate": 8.981073211314477e-06, | |
| "loss": 0.01, | |
| "step": 265000 | |
| }, | |
| { | |
| "epoch": 110.44, | |
| "learning_rate": 8.960316139767055e-06, | |
| "loss": 0.0099, | |
| "step": 265500 | |
| }, | |
| { | |
| "epoch": 110.65, | |
| "learning_rate": 8.939517470881864e-06, | |
| "loss": 0.0108, | |
| "step": 266000 | |
| }, | |
| { | |
| "epoch": 110.86, | |
| "learning_rate": 8.918718801996674e-06, | |
| "loss": 0.0109, | |
| "step": 266500 | |
| }, | |
| { | |
| "epoch": 111.0, | |
| "eval_cer": 0.2476, | |
| "eval_gen_len": 13.6859, | |
| "eval_loss": 0.4497167766094208, | |
| "eval_runtime": 326.8835, | |
| "eval_samples_per_second": 9.808, | |
| "eval_steps_per_second": 2.453, | |
| "step": 266844 | |
| }, | |
| { | |
| "epoch": 111.06, | |
| "learning_rate": 8.897920133111481e-06, | |
| "loss": 0.0106, | |
| "step": 267000 | |
| }, | |
| { | |
| "epoch": 111.27, | |
| "learning_rate": 8.87712146422629e-06, | |
| "loss": 0.0103, | |
| "step": 267500 | |
| }, | |
| { | |
| "epoch": 111.48, | |
| "learning_rate": 8.856322795341098e-06, | |
| "loss": 0.0104, | |
| "step": 268000 | |
| }, | |
| { | |
| "epoch": 111.69, | |
| "learning_rate": 8.835524126455908e-06, | |
| "loss": 0.0126, | |
| "step": 268500 | |
| }, | |
| { | |
| "epoch": 111.9, | |
| "learning_rate": 8.814725457570717e-06, | |
| "loss": 0.0115, | |
| "step": 269000 | |
| }, | |
| { | |
| "epoch": 112.0, | |
| "eval_cer": 0.2528, | |
| "eval_gen_len": 13.9145, | |
| "eval_loss": 0.4540727734565735, | |
| "eval_runtime": 317.49, | |
| "eval_samples_per_second": 10.098, | |
| "eval_steps_per_second": 2.526, | |
| "step": 269248 | |
| }, | |
| { | |
| "epoch": 112.1, | |
| "learning_rate": 8.794009983361066e-06, | |
| "loss": 0.0108, | |
| "step": 269500 | |
| }, | |
| { | |
| "epoch": 112.31, | |
| "learning_rate": 8.773211314475875e-06, | |
| "loss": 0.0092, | |
| "step": 270000 | |
| }, | |
| { | |
| "epoch": 112.52, | |
| "learning_rate": 8.752412645590683e-06, | |
| "loss": 0.0104, | |
| "step": 270500 | |
| }, | |
| { | |
| "epoch": 112.73, | |
| "learning_rate": 8.731613976705492e-06, | |
| "loss": 0.0091, | |
| "step": 271000 | |
| }, | |
| { | |
| "epoch": 112.94, | |
| "learning_rate": 8.7108153078203e-06, | |
| "loss": 0.0115, | |
| "step": 271500 | |
| }, | |
| { | |
| "epoch": 113.0, | |
| "eval_cer": 0.2449, | |
| "eval_gen_len": 13.4545, | |
| "eval_loss": 0.4440736770629883, | |
| "eval_runtime": 280.5689, | |
| "eval_samples_per_second": 11.427, | |
| "eval_steps_per_second": 2.858, | |
| "step": 271652 | |
| }, | |
| { | |
| "epoch": 113.14, | |
| "learning_rate": 8.69001663893511e-06, | |
| "loss": 0.0092, | |
| "step": 272000 | |
| }, | |
| { | |
| "epoch": 113.35, | |
| "learning_rate": 8.669217970049919e-06, | |
| "loss": 0.0092, | |
| "step": 272500 | |
| }, | |
| { | |
| "epoch": 113.56, | |
| "learning_rate": 8.648419301164726e-06, | |
| "loss": 0.0109, | |
| "step": 273000 | |
| }, | |
| { | |
| "epoch": 113.77, | |
| "learning_rate": 8.627620632279536e-06, | |
| "loss": 0.0098, | |
| "step": 273500 | |
| }, | |
| { | |
| "epoch": 113.98, | |
| "learning_rate": 8.606821963394343e-06, | |
| "loss": 0.0113, | |
| "step": 274000 | |
| }, | |
| { | |
| "epoch": 114.0, | |
| "eval_cer": 0.2471, | |
| "eval_gen_len": 13.5477, | |
| "eval_loss": 0.4469629228115082, | |
| "eval_runtime": 276.6766, | |
| "eval_samples_per_second": 11.588, | |
| "eval_steps_per_second": 2.899, | |
| "step": 274056 | |
| }, | |
| { | |
| "epoch": 114.18, | |
| "learning_rate": 8.586064891846923e-06, | |
| "loss": 0.0115, | |
| "step": 274500 | |
| }, | |
| { | |
| "epoch": 114.39, | |
| "learning_rate": 8.565266222961731e-06, | |
| "loss": 0.0077, | |
| "step": 275000 | |
| }, | |
| { | |
| "epoch": 114.6, | |
| "learning_rate": 8.54446755407654e-06, | |
| "loss": 0.0103, | |
| "step": 275500 | |
| }, | |
| { | |
| "epoch": 114.81, | |
| "learning_rate": 8.523668885191348e-06, | |
| "loss": 0.01, | |
| "step": 276000 | |
| }, | |
| { | |
| "epoch": 115.0, | |
| "eval_cer": 0.2537, | |
| "eval_gen_len": 13.9704, | |
| "eval_loss": 0.47477516531944275, | |
| "eval_runtime": 291.6272, | |
| "eval_samples_per_second": 10.993, | |
| "eval_steps_per_second": 2.75, | |
| "step": 276460 | |
| }, | |
| { | |
| "epoch": 115.02, | |
| "learning_rate": 8.502911813643926e-06, | |
| "loss": 0.0116, | |
| "step": 276500 | |
| }, | |
| { | |
| "epoch": 115.22, | |
| "learning_rate": 8.482113144758736e-06, | |
| "loss": 0.0102, | |
| "step": 277000 | |
| }, | |
| { | |
| "epoch": 115.43, | |
| "learning_rate": 8.461314475873545e-06, | |
| "loss": 0.0085, | |
| "step": 277500 | |
| }, | |
| { | |
| "epoch": 115.64, | |
| "learning_rate": 8.440515806988353e-06, | |
| "loss": 0.0104, | |
| "step": 278000 | |
| }, | |
| { | |
| "epoch": 115.85, | |
| "learning_rate": 8.419758735440932e-06, | |
| "loss": 0.0125, | |
| "step": 278500 | |
| }, | |
| { | |
| "epoch": 116.0, | |
| "eval_cer": 0.2438, | |
| "eval_gen_len": 13.4816, | |
| "eval_loss": 0.4395754933357239, | |
| "eval_runtime": 273.4039, | |
| "eval_samples_per_second": 11.726, | |
| "eval_steps_per_second": 2.933, | |
| "step": 278864 | |
| }, | |
| { | |
| "epoch": 116.06, | |
| "learning_rate": 8.398960066555742e-06, | |
| "loss": 0.0093, | |
| "step": 279000 | |
| }, | |
| { | |
| "epoch": 116.26, | |
| "learning_rate": 8.37816139767055e-06, | |
| "loss": 0.01, | |
| "step": 279500 | |
| }, | |
| { | |
| "epoch": 116.47, | |
| "learning_rate": 8.357362728785359e-06, | |
| "loss": 0.0087, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 116.68, | |
| "learning_rate": 8.336564059900167e-06, | |
| "loss": 0.0099, | |
| "step": 280500 | |
| }, | |
| { | |
| "epoch": 116.89, | |
| "learning_rate": 8.315765391014976e-06, | |
| "loss": 0.0101, | |
| "step": 281000 | |
| }, | |
| { | |
| "epoch": 117.0, | |
| "eval_cer": 0.2464, | |
| "eval_gen_len": 13.7723, | |
| "eval_loss": 0.44487103819847107, | |
| "eval_runtime": 278.0443, | |
| "eval_samples_per_second": 11.531, | |
| "eval_steps_per_second": 2.884, | |
| "step": 281268 | |
| }, | |
| { | |
| "epoch": 117.1, | |
| "learning_rate": 8.295008319467554e-06, | |
| "loss": 0.0094, | |
| "step": 281500 | |
| }, | |
| { | |
| "epoch": 117.3, | |
| "learning_rate": 8.274209650582364e-06, | |
| "loss": 0.0077, | |
| "step": 282000 | |
| }, | |
| { | |
| "epoch": 117.51, | |
| "learning_rate": 8.253410981697171e-06, | |
| "loss": 0.01, | |
| "step": 282500 | |
| }, | |
| { | |
| "epoch": 117.72, | |
| "learning_rate": 8.23261231281198e-06, | |
| "loss": 0.0113, | |
| "step": 283000 | |
| }, | |
| { | |
| "epoch": 117.93, | |
| "learning_rate": 8.21181364392679e-06, | |
| "loss": 0.0108, | |
| "step": 283500 | |
| }, | |
| { | |
| "epoch": 118.0, | |
| "eval_cer": 0.249, | |
| "eval_gen_len": 13.7711, | |
| "eval_loss": 0.4563674330711365, | |
| "eval_runtime": 286.5388, | |
| "eval_samples_per_second": 11.189, | |
| "eval_steps_per_second": 2.799, | |
| "step": 283672 | |
| }, | |
| { | |
| "epoch": 118.14, | |
| "learning_rate": 8.191014975041598e-06, | |
| "loss": 0.0093, | |
| "step": 284000 | |
| }, | |
| { | |
| "epoch": 118.34, | |
| "learning_rate": 8.170216306156407e-06, | |
| "loss": 0.0097, | |
| "step": 284500 | |
| }, | |
| { | |
| "epoch": 118.55, | |
| "learning_rate": 8.149459234608985e-06, | |
| "loss": 0.0088, | |
| "step": 285000 | |
| }, | |
| { | |
| "epoch": 118.76, | |
| "learning_rate": 8.128660565723795e-06, | |
| "loss": 0.0087, | |
| "step": 285500 | |
| }, | |
| { | |
| "epoch": 118.97, | |
| "learning_rate": 8.107861896838602e-06, | |
| "loss": 0.0121, | |
| "step": 286000 | |
| }, | |
| { | |
| "epoch": 119.0, | |
| "eval_cer": 0.2484, | |
| "eval_gen_len": 13.7848, | |
| "eval_loss": 0.4589692950248718, | |
| "eval_runtime": 285.9698, | |
| "eval_samples_per_second": 11.211, | |
| "eval_steps_per_second": 2.804, | |
| "step": 286076 | |
| }, | |
| { | |
| "epoch": 119.18, | |
| "learning_rate": 8.087063227953412e-06, | |
| "loss": 0.0092, | |
| "step": 286500 | |
| }, | |
| { | |
| "epoch": 119.38, | |
| "learning_rate": 8.06626455906822e-06, | |
| "loss": 0.0104, | |
| "step": 287000 | |
| }, | |
| { | |
| "epoch": 119.59, | |
| "learning_rate": 8.045465890183029e-06, | |
| "loss": 0.0094, | |
| "step": 287500 | |
| }, | |
| { | |
| "epoch": 119.8, | |
| "learning_rate": 8.024667221297838e-06, | |
| "loss": 0.0111, | |
| "step": 288000 | |
| }, | |
| { | |
| "epoch": 120.0, | |
| "eval_cer": 0.2488, | |
| "eval_gen_len": 13.8621, | |
| "eval_loss": 0.46863117814064026, | |
| "eval_runtime": 282.0826, | |
| "eval_samples_per_second": 11.365, | |
| "eval_steps_per_second": 2.843, | |
| "step": 288480 | |
| }, | |
| { | |
| "epoch": 120.01, | |
| "learning_rate": 8.003868552412646e-06, | |
| "loss": 0.0105, | |
| "step": 288500 | |
| }, | |
| { | |
| "epoch": 120.22, | |
| "learning_rate": 7.983111480865226e-06, | |
| "loss": 0.0086, | |
| "step": 289000 | |
| }, | |
| { | |
| "epoch": 120.42, | |
| "learning_rate": 7.962312811980035e-06, | |
| "loss": 0.0101, | |
| "step": 289500 | |
| }, | |
| { | |
| "epoch": 120.63, | |
| "learning_rate": 7.941514143094843e-06, | |
| "loss": 0.0103, | |
| "step": 290000 | |
| }, | |
| { | |
| "epoch": 120.84, | |
| "learning_rate": 7.920715474209652e-06, | |
| "loss": 0.009, | |
| "step": 290500 | |
| }, | |
| { | |
| "epoch": 121.0, | |
| "eval_cer": 0.2451, | |
| "eval_gen_len": 13.7648, | |
| "eval_loss": 0.44192788004875183, | |
| "eval_runtime": 276.6356, | |
| "eval_samples_per_second": 11.589, | |
| "eval_steps_per_second": 2.899, | |
| "step": 290884 | |
| }, | |
| { | |
| "epoch": 121.05, | |
| "learning_rate": 7.89991680532446e-06, | |
| "loss": 0.0103, | |
| "step": 291000 | |
| }, | |
| { | |
| "epoch": 121.26, | |
| "learning_rate": 7.87915973377704e-06, | |
| "loss": 0.0109, | |
| "step": 291500 | |
| }, | |
| { | |
| "epoch": 121.46, | |
| "learning_rate": 7.858361064891847e-06, | |
| "loss": 0.0096, | |
| "step": 292000 | |
| }, | |
| { | |
| "epoch": 121.67, | |
| "learning_rate": 7.837603993344427e-06, | |
| "loss": 0.0104, | |
| "step": 292500 | |
| }, | |
| { | |
| "epoch": 121.88, | |
| "learning_rate": 7.816805324459236e-06, | |
| "loss": 0.0093, | |
| "step": 293000 | |
| }, | |
| { | |
| "epoch": 122.0, | |
| "eval_cer": 0.242, | |
| "eval_gen_len": 13.6089, | |
| "eval_loss": 0.43688440322875977, | |
| "eval_runtime": 279.7745, | |
| "eval_samples_per_second": 11.459, | |
| "eval_steps_per_second": 2.867, | |
| "step": 293288 | |
| }, | |
| { | |
| "epoch": 122.09, | |
| "learning_rate": 7.796006655574044e-06, | |
| "loss": 0.0096, | |
| "step": 293500 | |
| }, | |
| { | |
| "epoch": 122.3, | |
| "learning_rate": 7.775207986688853e-06, | |
| "loss": 0.0083, | |
| "step": 294000 | |
| }, | |
| { | |
| "epoch": 122.5, | |
| "learning_rate": 7.754409317803661e-06, | |
| "loss": 0.0092, | |
| "step": 294500 | |
| }, | |
| { | |
| "epoch": 122.71, | |
| "learning_rate": 7.73361064891847e-06, | |
| "loss": 0.0119, | |
| "step": 295000 | |
| }, | |
| { | |
| "epoch": 122.92, | |
| "learning_rate": 7.71281198003328e-06, | |
| "loss": 0.0092, | |
| "step": 295500 | |
| }, | |
| { | |
| "epoch": 123.0, | |
| "eval_cer": 0.2475, | |
| "eval_gen_len": 13.6304, | |
| "eval_loss": 0.4489113390445709, | |
| "eval_runtime": 292.1033, | |
| "eval_samples_per_second": 10.976, | |
| "eval_steps_per_second": 2.746, | |
| "step": 295692 | |
| }, | |
| { | |
| "epoch": 123.13, | |
| "learning_rate": 7.692054908485858e-06, | |
| "loss": 0.0096, | |
| "step": 296000 | |
| }, | |
| { | |
| "epoch": 123.34, | |
| "learning_rate": 7.671256239600666e-06, | |
| "loss": 0.0093, | |
| "step": 296500 | |
| }, | |
| { | |
| "epoch": 123.54, | |
| "learning_rate": 7.650457570715475e-06, | |
| "loss": 0.0094, | |
| "step": 297000 | |
| }, | |
| { | |
| "epoch": 123.75, | |
| "learning_rate": 7.629658901830283e-06, | |
| "loss": 0.0111, | |
| "step": 297500 | |
| }, | |
| { | |
| "epoch": 123.96, | |
| "learning_rate": 7.608860232945092e-06, | |
| "loss": 0.0099, | |
| "step": 298000 | |
| }, | |
| { | |
| "epoch": 124.0, | |
| "eval_cer": 0.2447, | |
| "eval_gen_len": 13.6413, | |
| "eval_loss": 0.45137402415275574, | |
| "eval_runtime": 273.6913, | |
| "eval_samples_per_second": 11.714, | |
| "eval_steps_per_second": 2.93, | |
| "step": 298096 | |
| }, | |
| { | |
| "epoch": 124.17, | |
| "learning_rate": 7.588061564059901e-06, | |
| "loss": 0.0078, | |
| "step": 298500 | |
| }, | |
| { | |
| "epoch": 124.38, | |
| "learning_rate": 7.56726289517471e-06, | |
| "loss": 0.0093, | |
| "step": 299000 | |
| }, | |
| { | |
| "epoch": 124.58, | |
| "learning_rate": 7.5464642262895185e-06, | |
| "loss": 0.0089, | |
| "step": 299500 | |
| }, | |
| { | |
| "epoch": 124.79, | |
| "learning_rate": 7.525707154742097e-06, | |
| "loss": 0.01, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 125.0, | |
| "learning_rate": 7.5049500831946766e-06, | |
| "loss": 0.0103, | |
| "step": 300500 | |
| }, | |
| { | |
| "epoch": 125.0, | |
| "eval_cer": 0.2467, | |
| "eval_gen_len": 13.8041, | |
| "eval_loss": 0.45419880747795105, | |
| "eval_runtime": 277.7542, | |
| "eval_samples_per_second": 11.543, | |
| "eval_steps_per_second": 2.887, | |
| "step": 300500 | |
| }, | |
| { | |
| "epoch": 125.21, | |
| "learning_rate": 7.484151414309485e-06, | |
| "loss": 0.0076, | |
| "step": 301000 | |
| }, | |
| { | |
| "epoch": 125.42, | |
| "learning_rate": 7.463352745424294e-06, | |
| "loss": 0.0089, | |
| "step": 301500 | |
| }, | |
| { | |
| "epoch": 125.62, | |
| "learning_rate": 7.442554076539102e-06, | |
| "loss": 0.0089, | |
| "step": 302000 | |
| }, | |
| { | |
| "epoch": 125.83, | |
| "learning_rate": 7.4217554076539115e-06, | |
| "loss": 0.0121, | |
| "step": 302500 | |
| }, | |
| { | |
| "epoch": 126.0, | |
| "eval_cer": 0.2496, | |
| "eval_gen_len": 13.8525, | |
| "eval_loss": 0.4686892330646515, | |
| "eval_runtime": 288.7714, | |
| "eval_samples_per_second": 11.102, | |
| "eval_steps_per_second": 2.777, | |
| "step": 302904 | |
| }, | |
| { | |
| "epoch": 126.04, | |
| "learning_rate": 7.40095673876872e-06, | |
| "loss": 0.0091, | |
| "step": 303000 | |
| }, | |
| { | |
| "epoch": 126.25, | |
| "learning_rate": 7.3801580698835285e-06, | |
| "loss": 0.0089, | |
| "step": 303500 | |
| }, | |
| { | |
| "epoch": 126.46, | |
| "learning_rate": 7.359359400998337e-06, | |
| "loss": 0.0092, | |
| "step": 304000 | |
| }, | |
| { | |
| "epoch": 126.66, | |
| "learning_rate": 7.338560732113146e-06, | |
| "loss": 0.0081, | |
| "step": 304500 | |
| }, | |
| { | |
| "epoch": 126.87, | |
| "learning_rate": 7.317845257903495e-06, | |
| "loss": 0.0116, | |
| "step": 305000 | |
| }, | |
| { | |
| "epoch": 127.0, | |
| "eval_cer": 0.2443, | |
| "eval_gen_len": 13.6432, | |
| "eval_loss": 0.4484730660915375, | |
| "eval_runtime": 273.2803, | |
| "eval_samples_per_second": 11.732, | |
| "eval_steps_per_second": 2.935, | |
| "step": 305308 | |
| }, | |
| { | |
| "epoch": 127.08, | |
| "learning_rate": 7.297046589018303e-06, | |
| "loss": 0.0077, | |
| "step": 305500 | |
| }, | |
| { | |
| "epoch": 127.29, | |
| "learning_rate": 7.276247920133111e-06, | |
| "loss": 0.009, | |
| "step": 306000 | |
| }, | |
| { | |
| "epoch": 127.5, | |
| "learning_rate": 7.25544925124792e-06, | |
| "loss": 0.0094, | |
| "step": 306500 | |
| }, | |
| { | |
| "epoch": 127.7, | |
| "learning_rate": 7.234650582362729e-06, | |
| "loss": 0.0081, | |
| "step": 307000 | |
| }, | |
| { | |
| "epoch": 127.91, | |
| "learning_rate": 7.213851913477538e-06, | |
| "loss": 0.0105, | |
| "step": 307500 | |
| }, | |
| { | |
| "epoch": 128.0, | |
| "eval_cer": 0.2437, | |
| "eval_gen_len": 13.7661, | |
| "eval_loss": 0.4494189918041229, | |
| "eval_runtime": 282.0309, | |
| "eval_samples_per_second": 11.368, | |
| "eval_steps_per_second": 2.844, | |
| "step": 307712 | |
| }, | |
| { | |
| "epoch": 128.12, | |
| "learning_rate": 7.193094841930118e-06, | |
| "loss": 0.0085, | |
| "step": 308000 | |
| }, | |
| { | |
| "epoch": 128.33, | |
| "learning_rate": 7.172296173044926e-06, | |
| "loss": 0.0086, | |
| "step": 308500 | |
| }, | |
| { | |
| "epoch": 128.54, | |
| "learning_rate": 7.151497504159735e-06, | |
| "loss": 0.0091, | |
| "step": 309000 | |
| }, | |
| { | |
| "epoch": 128.74, | |
| "learning_rate": 7.130698835274543e-06, | |
| "loss": 0.0094, | |
| "step": 309500 | |
| }, | |
| { | |
| "epoch": 128.95, | |
| "learning_rate": 7.109900166389352e-06, | |
| "loss": 0.0087, | |
| "step": 310000 | |
| }, | |
| { | |
| "epoch": 129.0, | |
| "eval_cer": 0.2465, | |
| "eval_gen_len": 13.5352, | |
| "eval_loss": 0.46537643671035767, | |
| "eval_runtime": 272.0294, | |
| "eval_samples_per_second": 11.785, | |
| "eval_steps_per_second": 2.948, | |
| "step": 310116 | |
| }, | |
| { | |
| "epoch": 129.16, | |
| "learning_rate": 7.08910149750416e-06, | |
| "loss": 0.0088, | |
| "step": 310500 | |
| }, | |
| { | |
| "epoch": 129.37, | |
| "learning_rate": 7.0683028286189696e-06, | |
| "loss": 0.0076, | |
| "step": 311000 | |
| }, | |
| { | |
| "epoch": 129.58, | |
| "learning_rate": 7.047504159733778e-06, | |
| "loss": 0.0088, | |
| "step": 311500 | |
| }, | |
| { | |
| "epoch": 129.78, | |
| "learning_rate": 7.026705490848587e-06, | |
| "loss": 0.0092, | |
| "step": 312000 | |
| }, | |
| { | |
| "epoch": 129.99, | |
| "learning_rate": 7.005948419301165e-06, | |
| "loss": 0.0106, | |
| "step": 312500 | |
| }, | |
| { | |
| "epoch": 130.0, | |
| "eval_cer": 0.2457, | |
| "eval_gen_len": 13.5667, | |
| "eval_loss": 0.44364768266677856, | |
| "eval_runtime": 269.0291, | |
| "eval_samples_per_second": 11.917, | |
| "eval_steps_per_second": 2.981, | |
| "step": 312520 | |
| }, | |
| { | |
| "epoch": 130.2, | |
| "learning_rate": 6.985149750415974e-06, | |
| "loss": 0.008, | |
| "step": 313000 | |
| }, | |
| { | |
| "epoch": 130.41, | |
| "learning_rate": 6.964351081530783e-06, | |
| "loss": 0.0072, | |
| "step": 313500 | |
| }, | |
| { | |
| "epoch": 130.62, | |
| "learning_rate": 6.943552412645591e-06, | |
| "loss": 0.0091, | |
| "step": 314000 | |
| }, | |
| { | |
| "epoch": 130.82, | |
| "learning_rate": 6.9227537437604e-06, | |
| "loss": 0.0087, | |
| "step": 314500 | |
| }, | |
| { | |
| "epoch": 131.0, | |
| "eval_cer": 0.2451, | |
| "eval_gen_len": 13.7358, | |
| "eval_loss": 0.4612971544265747, | |
| "eval_runtime": 274.1913, | |
| "eval_samples_per_second": 11.693, | |
| "eval_steps_per_second": 2.925, | |
| "step": 314924 | |
| }, | |
| { | |
| "epoch": 131.03, | |
| "learning_rate": 6.90199667221298e-06, | |
| "loss": 0.0107, | |
| "step": 315000 | |
| }, | |
| { | |
| "epoch": 131.24, | |
| "learning_rate": 6.881239600665558e-06, | |
| "loss": 0.0082, | |
| "step": 315500 | |
| }, | |
| { | |
| "epoch": 131.45, | |
| "learning_rate": 6.860440931780366e-06, | |
| "loss": 0.0089, | |
| "step": 316000 | |
| }, | |
| { | |
| "epoch": 131.66, | |
| "learning_rate": 6.839642262895176e-06, | |
| "loss": 0.009, | |
| "step": 316500 | |
| }, | |
| { | |
| "epoch": 131.86, | |
| "learning_rate": 6.818843594009984e-06, | |
| "loss": 0.0104, | |
| "step": 317000 | |
| }, | |
| { | |
| "epoch": 132.0, | |
| "eval_cer": 0.2468, | |
| "eval_gen_len": 13.5936, | |
| "eval_loss": 0.4653697907924652, | |
| "eval_runtime": 277.9546, | |
| "eval_samples_per_second": 11.534, | |
| "eval_steps_per_second": 2.885, | |
| "step": 317328 | |
| }, | |
| { | |
| "epoch": 132.07, | |
| "learning_rate": 6.798044925124793e-06, | |
| "loss": 0.0094, | |
| "step": 317500 | |
| }, | |
| { | |
| "epoch": 132.28, | |
| "learning_rate": 6.777246256239601e-06, | |
| "loss": 0.0079, | |
| "step": 318000 | |
| }, | |
| { | |
| "epoch": 132.49, | |
| "learning_rate": 6.75644758735441e-06, | |
| "loss": 0.0098, | |
| "step": 318500 | |
| }, | |
| { | |
| "epoch": 132.7, | |
| "learning_rate": 6.735648918469218e-06, | |
| "loss": 0.0081, | |
| "step": 319000 | |
| }, | |
| { | |
| "epoch": 132.9, | |
| "learning_rate": 6.714891846921797e-06, | |
| "loss": 0.0089, | |
| "step": 319500 | |
| }, | |
| { | |
| "epoch": 133.0, | |
| "eval_cer": 0.2455, | |
| "eval_gen_len": 13.6875, | |
| "eval_loss": 0.45620593428611755, | |
| "eval_runtime": 284.3664, | |
| "eval_samples_per_second": 11.274, | |
| "eval_steps_per_second": 2.82, | |
| "step": 319732 | |
| }, | |
| { | |
| "epoch": 133.11, | |
| "learning_rate": 6.694093178036606e-06, | |
| "loss": 0.0085, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 133.32, | |
| "learning_rate": 6.673294509151414e-06, | |
| "loss": 0.0091, | |
| "step": 320500 | |
| }, | |
| { | |
| "epoch": 133.53, | |
| "learning_rate": 6.652495840266223e-06, | |
| "loss": 0.0073, | |
| "step": 321000 | |
| }, | |
| { | |
| "epoch": 133.74, | |
| "learning_rate": 6.631738768718803e-06, | |
| "loss": 0.0091, | |
| "step": 321500 | |
| }, | |
| { | |
| "epoch": 133.94, | |
| "learning_rate": 6.610940099833611e-06, | |
| "loss": 0.0088, | |
| "step": 322000 | |
| }, | |
| { | |
| "epoch": 134.0, | |
| "eval_cer": 0.2455, | |
| "eval_gen_len": 13.6653, | |
| "eval_loss": 0.4463290274143219, | |
| "eval_runtime": 265.7595, | |
| "eval_samples_per_second": 12.064, | |
| "eval_steps_per_second": 3.018, | |
| "step": 322136 | |
| }, | |
| { | |
| "epoch": 134.15, | |
| "learning_rate": 6.59014143094842e-06, | |
| "loss": 0.0082, | |
| "step": 322500 | |
| }, | |
| { | |
| "epoch": 134.36, | |
| "learning_rate": 6.569342762063229e-06, | |
| "loss": 0.0087, | |
| "step": 323000 | |
| }, | |
| { | |
| "epoch": 134.57, | |
| "learning_rate": 6.548544093178038e-06, | |
| "loss": 0.009, | |
| "step": 323500 | |
| }, | |
| { | |
| "epoch": 134.78, | |
| "learning_rate": 6.527745424292846e-06, | |
| "loss": 0.0093, | |
| "step": 324000 | |
| }, | |
| { | |
| "epoch": 134.98, | |
| "learning_rate": 6.506946755407655e-06, | |
| "loss": 0.0088, | |
| "step": 324500 | |
| }, | |
| { | |
| "epoch": 135.0, | |
| "eval_cer": 0.245, | |
| "eval_gen_len": 13.5749, | |
| "eval_loss": 0.4489509165287018, | |
| "eval_runtime": 256.739, | |
| "eval_samples_per_second": 12.487, | |
| "eval_steps_per_second": 3.124, | |
| "step": 324540 | |
| }, | |
| { | |
| "epoch": 135.19, | |
| "learning_rate": 6.486148086522463e-06, | |
| "loss": 0.0076, | |
| "step": 325000 | |
| }, | |
| { | |
| "epoch": 135.4, | |
| "learning_rate": 6.465349417637273e-06, | |
| "loss": 0.0078, | |
| "step": 325500 | |
| }, | |
| { | |
| "epoch": 135.61, | |
| "learning_rate": 6.444550748752081e-06, | |
| "loss": 0.007, | |
| "step": 326000 | |
| }, | |
| { | |
| "epoch": 135.82, | |
| "learning_rate": 6.423793677204659e-06, | |
| "loss": 0.0086, | |
| "step": 326500 | |
| }, | |
| { | |
| "epoch": 136.0, | |
| "eval_cer": 0.2458, | |
| "eval_gen_len": 13.8032, | |
| "eval_loss": 0.45637834072113037, | |
| "eval_runtime": 275.2079, | |
| "eval_samples_per_second": 11.649, | |
| "eval_steps_per_second": 2.914, | |
| "step": 326944 | |
| }, | |
| { | |
| "epoch": 136.02, | |
| "learning_rate": 6.402995008319468e-06, | |
| "loss": 0.0097, | |
| "step": 327000 | |
| }, | |
| { | |
| "epoch": 136.23, | |
| "learning_rate": 6.382237936772047e-06, | |
| "loss": 0.0084, | |
| "step": 327500 | |
| }, | |
| { | |
| "epoch": 136.44, | |
| "learning_rate": 6.361439267886855e-06, | |
| "loss": 0.0084, | |
| "step": 328000 | |
| }, | |
| { | |
| "epoch": 136.65, | |
| "learning_rate": 6.340640599001664e-06, | |
| "loss": 0.0083, | |
| "step": 328500 | |
| }, | |
| { | |
| "epoch": 136.86, | |
| "learning_rate": 6.319841930116472e-06, | |
| "loss": 0.0083, | |
| "step": 329000 | |
| }, | |
| { | |
| "epoch": 137.0, | |
| "eval_cer": 0.2471, | |
| "eval_gen_len": 13.6478, | |
| "eval_loss": 0.4573554992675781, | |
| "eval_runtime": 269.5066, | |
| "eval_samples_per_second": 11.896, | |
| "eval_steps_per_second": 2.976, | |
| "step": 329348 | |
| }, | |
| { | |
| "epoch": 137.06, | |
| "learning_rate": 6.299043261231281e-06, | |
| "loss": 0.0089, | |
| "step": 329500 | |
| }, | |
| { | |
| "epoch": 137.27, | |
| "learning_rate": 6.27824459234609e-06, | |
| "loss": 0.0087, | |
| "step": 330000 | |
| }, | |
| { | |
| "epoch": 137.48, | |
| "learning_rate": 6.257445923460899e-06, | |
| "loss": 0.0086, | |
| "step": 330500 | |
| }, | |
| { | |
| "epoch": 137.69, | |
| "learning_rate": 6.236647254575707e-06, | |
| "loss": 0.0085, | |
| "step": 331000 | |
| }, | |
| { | |
| "epoch": 137.9, | |
| "learning_rate": 6.215890183028287e-06, | |
| "loss": 0.0092, | |
| "step": 331500 | |
| }, | |
| { | |
| "epoch": 138.0, | |
| "eval_cer": 0.2487, | |
| "eval_gen_len": 13.8503, | |
| "eval_loss": 0.46958354115486145, | |
| "eval_runtime": 265.4531, | |
| "eval_samples_per_second": 12.077, | |
| "eval_steps_per_second": 3.021, | |
| "step": 331752 | |
| }, | |
| { | |
| "epoch": 138.1, | |
| "learning_rate": 6.195091514143096e-06, | |
| "loss": 0.0079, | |
| "step": 332000 | |
| }, | |
| { | |
| "epoch": 138.31, | |
| "learning_rate": 6.174292845257904e-06, | |
| "loss": 0.0073, | |
| "step": 332500 | |
| }, | |
| { | |
| "epoch": 138.52, | |
| "learning_rate": 6.153494176372713e-06, | |
| "loss": 0.0074, | |
| "step": 333000 | |
| }, | |
| { | |
| "epoch": 138.73, | |
| "learning_rate": 6.132695507487521e-06, | |
| "loss": 0.0089, | |
| "step": 333500 | |
| }, | |
| { | |
| "epoch": 138.94, | |
| "learning_rate": 6.111896838602331e-06, | |
| "loss": 0.0082, | |
| "step": 334000 | |
| }, | |
| { | |
| "epoch": 139.0, | |
| "eval_cer": 0.2476, | |
| "eval_gen_len": 13.7916, | |
| "eval_loss": 0.4610365629196167, | |
| "eval_runtime": 273.061, | |
| "eval_samples_per_second": 11.741, | |
| "eval_steps_per_second": 2.937, | |
| "step": 334156 | |
| }, | |
| { | |
| "epoch": 139.14, | |
| "learning_rate": 6.091098169717139e-06, | |
| "loss": 0.0076, | |
| "step": 334500 | |
| }, | |
| { | |
| "epoch": 139.35, | |
| "learning_rate": 6.070299500831948e-06, | |
| "loss": 0.0083, | |
| "step": 335000 | |
| }, | |
| { | |
| "epoch": 139.56, | |
| "learning_rate": 6.049542429284526e-06, | |
| "loss": 0.0088, | |
| "step": 335500 | |
| }, | |
| { | |
| "epoch": 139.77, | |
| "learning_rate": 6.0287437603993344e-06, | |
| "loss": 0.0078, | |
| "step": 336000 | |
| }, | |
| { | |
| "epoch": 139.98, | |
| "learning_rate": 6.007945091514144e-06, | |
| "loss": 0.0092, | |
| "step": 336500 | |
| }, | |
| { | |
| "epoch": 140.0, | |
| "eval_cer": 0.2432, | |
| "eval_gen_len": 13.6344, | |
| "eval_loss": 0.43919724225997925, | |
| "eval_runtime": 265.4301, | |
| "eval_samples_per_second": 12.079, | |
| "eval_steps_per_second": 3.022, | |
| "step": 336560 | |
| }, | |
| { | |
| "epoch": 140.18, | |
| "learning_rate": 5.987146422628952e-06, | |
| "loss": 0.007, | |
| "step": 337000 | |
| }, | |
| { | |
| "epoch": 140.39, | |
| "learning_rate": 5.966347753743761e-06, | |
| "loss": 0.0078, | |
| "step": 337500 | |
| }, | |
| { | |
| "epoch": 140.6, | |
| "learning_rate": 5.945590682196339e-06, | |
| "loss": 0.0087, | |
| "step": 338000 | |
| }, | |
| { | |
| "epoch": 140.81, | |
| "learning_rate": 5.924792013311148e-06, | |
| "loss": 0.0083, | |
| "step": 338500 | |
| }, | |
| { | |
| "epoch": 141.0, | |
| "eval_cer": 0.2461, | |
| "eval_gen_len": 13.733, | |
| "eval_loss": 0.4848983883857727, | |
| "eval_runtime": 269.6844, | |
| "eval_samples_per_second": 11.888, | |
| "eval_steps_per_second": 2.974, | |
| "step": 338964 | |
| }, | |
| { | |
| "epoch": 141.01, | |
| "learning_rate": 5.904034941763727e-06, | |
| "loss": 0.0083, | |
| "step": 339000 | |
| }, | |
| { | |
| "epoch": 141.22, | |
| "learning_rate": 5.883236272878537e-06, | |
| "loss": 0.0079, | |
| "step": 339500 | |
| }, | |
| { | |
| "epoch": 141.43, | |
| "learning_rate": 5.862437603993345e-06, | |
| "loss": 0.007, | |
| "step": 340000 | |
| }, | |
| { | |
| "epoch": 141.64, | |
| "learning_rate": 5.841638935108154e-06, | |
| "loss": 0.0085, | |
| "step": 340500 | |
| }, | |
| { | |
| "epoch": 141.85, | |
| "learning_rate": 5.820840266222962e-06, | |
| "loss": 0.0085, | |
| "step": 341000 | |
| }, | |
| { | |
| "epoch": 142.0, | |
| "eval_cer": 0.2475, | |
| "eval_gen_len": 13.8278, | |
| "eval_loss": 0.46004167199134827, | |
| "eval_runtime": 262.9827, | |
| "eval_samples_per_second": 12.191, | |
| "eval_steps_per_second": 3.05, | |
| "step": 341368 | |
| }, | |
| { | |
| "epoch": 142.05, | |
| "learning_rate": 5.800041597337771e-06, | |
| "loss": 0.0079, | |
| "step": 341500 | |
| }, | |
| { | |
| "epoch": 142.26, | |
| "learning_rate": 5.779242928452579e-06, | |
| "loss": 0.0078, | |
| "step": 342000 | |
| }, | |
| { | |
| "epoch": 142.47, | |
| "learning_rate": 5.758444259567389e-06, | |
| "loss": 0.0083, | |
| "step": 342500 | |
| }, | |
| { | |
| "epoch": 142.68, | |
| "learning_rate": 5.737645590682197e-06, | |
| "loss": 0.0072, | |
| "step": 343000 | |
| }, | |
| { | |
| "epoch": 142.89, | |
| "learning_rate": 5.7168885191347755e-06, | |
| "loss": 0.008, | |
| "step": 343500 | |
| }, | |
| { | |
| "epoch": 143.0, | |
| "eval_cer": 0.2455, | |
| "eval_gen_len": 13.7137, | |
| "eval_loss": 0.4594569206237793, | |
| "eval_runtime": 270.269, | |
| "eval_samples_per_second": 11.862, | |
| "eval_steps_per_second": 2.967, | |
| "step": 343772 | |
| }, | |
| { | |
| "epoch": 143.09, | |
| "learning_rate": 5.696089850249584e-06, | |
| "loss": 0.0085, | |
| "step": 344000 | |
| }, | |
| { | |
| "epoch": 143.3, | |
| "learning_rate": 5.675291181364393e-06, | |
| "loss": 0.0082, | |
| "step": 344500 | |
| }, | |
| { | |
| "epoch": 143.51, | |
| "learning_rate": 5.654492512479202e-06, | |
| "loss": 0.0078, | |
| "step": 345000 | |
| }, | |
| { | |
| "epoch": 143.72, | |
| "learning_rate": 5.633735440931781e-06, | |
| "loss": 0.0091, | |
| "step": 345500 | |
| }, | |
| { | |
| "epoch": 143.93, | |
| "learning_rate": 5.61293677204659e-06, | |
| "loss": 0.0084, | |
| "step": 346000 | |
| }, | |
| { | |
| "epoch": 144.0, | |
| "eval_cer": 0.2419, | |
| "eval_gen_len": 13.684, | |
| "eval_loss": 0.44198158383369446, | |
| "eval_runtime": 272.3122, | |
| "eval_samples_per_second": 11.773, | |
| "eval_steps_per_second": 2.945, | |
| "step": 346176 | |
| }, | |
| { | |
| "epoch": 144.13, | |
| "learning_rate": 5.592138103161399e-06, | |
| "loss": 0.0073, | |
| "step": 346500 | |
| }, | |
| { | |
| "epoch": 144.34, | |
| "learning_rate": 5.571339434276207e-06, | |
| "loss": 0.0076, | |
| "step": 347000 | |
| }, | |
| { | |
| "epoch": 144.55, | |
| "learning_rate": 5.550540765391016e-06, | |
| "loss": 0.0075, | |
| "step": 347500 | |
| }, | |
| { | |
| "epoch": 144.76, | |
| "learning_rate": 5.529783693843595e-06, | |
| "loss": 0.0092, | |
| "step": 348000 | |
| }, | |
| { | |
| "epoch": 144.97, | |
| "learning_rate": 5.508985024958403e-06, | |
| "loss": 0.0087, | |
| "step": 348500 | |
| }, | |
| { | |
| "epoch": 145.0, | |
| "eval_cer": 0.2435, | |
| "eval_gen_len": 13.5817, | |
| "eval_loss": 0.4379993975162506, | |
| "eval_runtime": 266.6103, | |
| "eval_samples_per_second": 12.025, | |
| "eval_steps_per_second": 3.008, | |
| "step": 348580 | |
| }, | |
| { | |
| "epoch": 145.17, | |
| "learning_rate": 5.488186356073212e-06, | |
| "loss": 0.007, | |
| "step": 349000 | |
| }, | |
| { | |
| "epoch": 145.38, | |
| "learning_rate": 5.46738768718802e-06, | |
| "loss": 0.0071, | |
| "step": 349500 | |
| }, | |
| { | |
| "epoch": 145.59, | |
| "learning_rate": 5.4466306156405994e-06, | |
| "loss": 0.0086, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 145.8, | |
| "learning_rate": 5.425831946755408e-06, | |
| "loss": 0.0074, | |
| "step": 350500 | |
| }, | |
| { | |
| "epoch": 146.0, | |
| "eval_cer": 0.2438, | |
| "eval_gen_len": 13.747, | |
| "eval_loss": 0.4474620223045349, | |
| "eval_runtime": 259.9915, | |
| "eval_samples_per_second": 12.331, | |
| "eval_steps_per_second": 3.085, | |
| "step": 350984 | |
| }, | |
| { | |
| "epoch": 146.01, | |
| "learning_rate": 5.4050332778702165e-06, | |
| "loss": 0.0085, | |
| "step": 351000 | |
| }, | |
| { | |
| "epoch": 146.21, | |
| "learning_rate": 5.384234608985025e-06, | |
| "loss": 0.0071, | |
| "step": 351500 | |
| }, | |
| { | |
| "epoch": 146.42, | |
| "learning_rate": 5.3634359400998335e-06, | |
| "loss": 0.0065, | |
| "step": 352000 | |
| }, | |
| { | |
| "epoch": 146.63, | |
| "learning_rate": 5.342637271214642e-06, | |
| "loss": 0.0076, | |
| "step": 352500 | |
| }, | |
| { | |
| "epoch": 146.84, | |
| "learning_rate": 5.3218386023294514e-06, | |
| "loss": 0.0076, | |
| "step": 353000 | |
| }, | |
| { | |
| "epoch": 147.0, | |
| "eval_cer": 0.2433, | |
| "eval_gen_len": 13.5989, | |
| "eval_loss": 0.4507221579551697, | |
| "eval_runtime": 259.806, | |
| "eval_samples_per_second": 12.34, | |
| "eval_steps_per_second": 3.087, | |
| "step": 353388 | |
| }, | |
| { | |
| "epoch": 147.05, | |
| "learning_rate": 5.30103993344426e-06, | |
| "loss": 0.0078, | |
| "step": 353500 | |
| }, | |
| { | |
| "epoch": 147.25, | |
| "learning_rate": 5.2802412645590685e-06, | |
| "loss": 0.0077, | |
| "step": 354000 | |
| }, | |
| { | |
| "epoch": 147.46, | |
| "learning_rate": 5.259484193011648e-06, | |
| "loss": 0.0075, | |
| "step": 354500 | |
| }, | |
| { | |
| "epoch": 147.67, | |
| "learning_rate": 5.238685524126457e-06, | |
| "loss": 0.0068, | |
| "step": 355000 | |
| }, | |
| { | |
| "epoch": 147.88, | |
| "learning_rate": 5.217886855241265e-06, | |
| "loss": 0.0091, | |
| "step": 355500 | |
| }, | |
| { | |
| "epoch": 148.0, | |
| "eval_cer": 0.2461, | |
| "eval_gen_len": 13.8125, | |
| "eval_loss": 0.47150808572769165, | |
| "eval_runtime": 262.9368, | |
| "eval_samples_per_second": 12.193, | |
| "eval_steps_per_second": 3.05, | |
| "step": 355792 | |
| }, | |
| { | |
| "epoch": 148.09, | |
| "learning_rate": 5.1971297836938436e-06, | |
| "loss": 0.0082, | |
| "step": 356000 | |
| }, | |
| { | |
| "epoch": 148.29, | |
| "learning_rate": 5.176331114808653e-06, | |
| "loss": 0.0066, | |
| "step": 356500 | |
| }, | |
| { | |
| "epoch": 148.5, | |
| "learning_rate": 5.1555324459234614e-06, | |
| "loss": 0.0081, | |
| "step": 357000 | |
| }, | |
| { | |
| "epoch": 148.71, | |
| "learning_rate": 5.13473377703827e-06, | |
| "loss": 0.0074, | |
| "step": 357500 | |
| }, | |
| { | |
| "epoch": 148.92, | |
| "learning_rate": 5.1139351081530785e-06, | |
| "loss": 0.0078, | |
| "step": 358000 | |
| }, | |
| { | |
| "epoch": 149.0, | |
| "eval_cer": 0.2472, | |
| "eval_gen_len": 13.8253, | |
| "eval_loss": 0.46047914028167725, | |
| "eval_runtime": 271.5636, | |
| "eval_samples_per_second": 11.806, | |
| "eval_steps_per_second": 2.953, | |
| "step": 358196 | |
| }, | |
| { | |
| "epoch": 149.13, | |
| "learning_rate": 5.093136439267887e-06, | |
| "loss": 0.0076, | |
| "step": 358500 | |
| }, | |
| { | |
| "epoch": 149.33, | |
| "learning_rate": 5.0723377703826955e-06, | |
| "loss": 0.0084, | |
| "step": 359000 | |
| }, | |
| { | |
| "epoch": 149.54, | |
| "learning_rate": 5.051539101497505e-06, | |
| "loss": 0.0081, | |
| "step": 359500 | |
| }, | |
| { | |
| "epoch": 149.75, | |
| "learning_rate": 5.0307404326123134e-06, | |
| "loss": 0.0075, | |
| "step": 360000 | |
| }, | |
| { | |
| "epoch": 149.96, | |
| "learning_rate": 5.009941763727122e-06, | |
| "loss": 0.0075, | |
| "step": 360500 | |
| }, | |
| { | |
| "epoch": 150.0, | |
| "eval_cer": 0.2416, | |
| "eval_gen_len": 13.4729, | |
| "eval_loss": 0.4448852837085724, | |
| "eval_runtime": 262.3745, | |
| "eval_samples_per_second": 12.219, | |
| "eval_steps_per_second": 3.057, | |
| "step": 360600 | |
| }, | |
| { | |
| "epoch": 150.17, | |
| "learning_rate": 4.989184692179701e-06, | |
| "loss": 0.0078, | |
| "step": 361000 | |
| }, | |
| { | |
| "epoch": 150.37, | |
| "learning_rate": 4.9683860232945095e-06, | |
| "loss": 0.0065, | |
| "step": 361500 | |
| }, | |
| { | |
| "epoch": 150.58, | |
| "learning_rate": 4.947587354409318e-06, | |
| "loss": 0.0077, | |
| "step": 362000 | |
| }, | |
| { | |
| "epoch": 150.79, | |
| "learning_rate": 4.926788685524127e-06, | |
| "loss": 0.008, | |
| "step": 362500 | |
| }, | |
| { | |
| "epoch": 151.0, | |
| "learning_rate": 4.905990016638936e-06, | |
| "loss": 0.0076, | |
| "step": 363000 | |
| }, | |
| { | |
| "epoch": 151.0, | |
| "eval_cer": 0.2458, | |
| "eval_gen_len": 13.7838, | |
| "eval_loss": 0.47447100281715393, | |
| "eval_runtime": 262.0741, | |
| "eval_samples_per_second": 12.233, | |
| "eval_steps_per_second": 3.06, | |
| "step": 363004 | |
| }, | |
| { | |
| "epoch": 151.21, | |
| "learning_rate": 4.885232945091515e-06, | |
| "loss": 0.0075, | |
| "step": 363500 | |
| }, | |
| { | |
| "epoch": 151.41, | |
| "learning_rate": 4.8644342762063235e-06, | |
| "loss": 0.0074, | |
| "step": 364000 | |
| }, | |
| { | |
| "epoch": 151.62, | |
| "learning_rate": 4.843635607321132e-06, | |
| "loss": 0.0077, | |
| "step": 364500 | |
| }, | |
| { | |
| "epoch": 151.83, | |
| "learning_rate": 4.822878535773711e-06, | |
| "loss": 0.0076, | |
| "step": 365000 | |
| }, | |
| { | |
| "epoch": 152.0, | |
| "eval_cer": 0.2461, | |
| "eval_gen_len": 13.7093, | |
| "eval_loss": 0.4570690095424652, | |
| "eval_runtime": 264.709, | |
| "eval_samples_per_second": 12.111, | |
| "eval_steps_per_second": 3.03, | |
| "step": 365408 | |
| }, | |
| { | |
| "epoch": 152.04, | |
| "learning_rate": 4.8020798668885195e-06, | |
| "loss": 0.0075, | |
| "step": 365500 | |
| }, | |
| { | |
| "epoch": 152.25, | |
| "learning_rate": 4.781281198003328e-06, | |
| "loss": 0.0066, | |
| "step": 366000 | |
| }, | |
| { | |
| "epoch": 152.45, | |
| "learning_rate": 4.7604825291181366e-06, | |
| "loss": 0.0079, | |
| "step": 366500 | |
| }, | |
| { | |
| "epoch": 152.66, | |
| "learning_rate": 4.739683860232945e-06, | |
| "loss": 0.0076, | |
| "step": 367000 | |
| }, | |
| { | |
| "epoch": 152.87, | |
| "learning_rate": 4.718885191347754e-06, | |
| "loss": 0.0088, | |
| "step": 367500 | |
| }, | |
| { | |
| "epoch": 153.0, | |
| "eval_cer": 0.2414, | |
| "eval_gen_len": 13.5468, | |
| "eval_loss": 0.45958101749420166, | |
| "eval_runtime": 263.6085, | |
| "eval_samples_per_second": 12.162, | |
| "eval_steps_per_second": 3.042, | |
| "step": 367812 | |
| }, | |
| { | |
| "epoch": 153.08, | |
| "learning_rate": 4.6981281198003335e-06, | |
| "loss": 0.0081, | |
| "step": 368000 | |
| }, | |
| { | |
| "epoch": 153.29, | |
| "learning_rate": 4.677329450915142e-06, | |
| "loss": 0.0072, | |
| "step": 368500 | |
| }, | |
| { | |
| "epoch": 153.49, | |
| "learning_rate": 4.6565307820299505e-06, | |
| "loss": 0.007, | |
| "step": 369000 | |
| }, | |
| { | |
| "epoch": 153.7, | |
| "learning_rate": 4.635732113144759e-06, | |
| "loss": 0.0068, | |
| "step": 369500 | |
| }, | |
| { | |
| "epoch": 153.91, | |
| "learning_rate": 4.6149334442595676e-06, | |
| "loss": 0.0082, | |
| "step": 370000 | |
| }, | |
| { | |
| "epoch": 154.0, | |
| "eval_cer": 0.2475, | |
| "eval_gen_len": 13.8384, | |
| "eval_loss": 0.46518319845199585, | |
| "eval_runtime": 270.3025, | |
| "eval_samples_per_second": 11.861, | |
| "eval_steps_per_second": 2.967, | |
| "step": 370216 | |
| }, | |
| { | |
| "epoch": 154.12, | |
| "learning_rate": 4.594134775374376e-06, | |
| "loss": 0.0067, | |
| "step": 370500 | |
| }, | |
| { | |
| "epoch": 154.33, | |
| "learning_rate": 4.5733361064891855e-06, | |
| "loss": 0.0073, | |
| "step": 371000 | |
| }, | |
| { | |
| "epoch": 154.53, | |
| "learning_rate": 4.552537437603994e-06, | |
| "loss": 0.0083, | |
| "step": 371500 | |
| }, | |
| { | |
| "epoch": 154.74, | |
| "learning_rate": 4.5317387687188025e-06, | |
| "loss": 0.0074, | |
| "step": 372000 | |
| }, | |
| { | |
| "epoch": 154.95, | |
| "learning_rate": 4.510940099833611e-06, | |
| "loss": 0.0077, | |
| "step": 372500 | |
| }, | |
| { | |
| "epoch": 155.0, | |
| "eval_cer": 0.2426, | |
| "eval_gen_len": 13.6457, | |
| "eval_loss": 0.46483084559440613, | |
| "eval_runtime": 258.2815, | |
| "eval_samples_per_second": 12.413, | |
| "eval_steps_per_second": 3.105, | |
| "step": 372620 | |
| }, | |
| { | |
| "epoch": 155.16, | |
| "learning_rate": 4.49018302828619e-06, | |
| "loss": 0.0072, | |
| "step": 373000 | |
| }, | |
| { | |
| "epoch": 155.37, | |
| "learning_rate": 4.469384359400999e-06, | |
| "loss": 0.0066, | |
| "step": 373500 | |
| }, | |
| { | |
| "epoch": 155.57, | |
| "learning_rate": 4.448627287853578e-06, | |
| "loss": 0.0074, | |
| "step": 374000 | |
| }, | |
| { | |
| "epoch": 155.78, | |
| "learning_rate": 4.427828618968386e-06, | |
| "loss": 0.0075, | |
| "step": 374500 | |
| }, | |
| { | |
| "epoch": 155.99, | |
| "learning_rate": 4.407029950083195e-06, | |
| "loss": 0.0074, | |
| "step": 375000 | |
| }, | |
| { | |
| "epoch": 156.0, | |
| "eval_cer": 0.2422, | |
| "eval_gen_len": 13.5889, | |
| "eval_loss": 0.4521370232105255, | |
| "eval_runtime": 266.1373, | |
| "eval_samples_per_second": 12.046, | |
| "eval_steps_per_second": 3.013, | |
| "step": 375024 | |
| }, | |
| { | |
| "epoch": 156.2, | |
| "learning_rate": 4.386231281198003e-06, | |
| "loss": 0.0072, | |
| "step": 375500 | |
| }, | |
| { | |
| "epoch": 156.41, | |
| "learning_rate": 4.3654326123128125e-06, | |
| "loss": 0.0073, | |
| "step": 376000 | |
| }, | |
| { | |
| "epoch": 156.61, | |
| "learning_rate": 4.344633943427621e-06, | |
| "loss": 0.0076, | |
| "step": 376500 | |
| }, | |
| { | |
| "epoch": 156.82, | |
| "learning_rate": 4.32383527454243e-06, | |
| "loss": 0.0073, | |
| "step": 377000 | |
| }, | |
| { | |
| "epoch": 157.0, | |
| "eval_cer": 0.2465, | |
| "eval_gen_len": 13.7867, | |
| "eval_loss": 0.47169268131256104, | |
| "eval_runtime": 264.268, | |
| "eval_samples_per_second": 12.132, | |
| "eval_steps_per_second": 3.035, | |
| "step": 377428 | |
| }, | |
| { | |
| "epoch": 157.03, | |
| "learning_rate": 4.303036605657238e-06, | |
| "loss": 0.0075, | |
| "step": 377500 | |
| }, | |
| { | |
| "epoch": 157.24, | |
| "learning_rate": 4.282237936772047e-06, | |
| "loss": 0.0063, | |
| "step": 378000 | |
| }, | |
| { | |
| "epoch": 157.45, | |
| "learning_rate": 4.261480865224626e-06, | |
| "loss": 0.0069, | |
| "step": 378500 | |
| }, | |
| { | |
| "epoch": 157.65, | |
| "learning_rate": 4.240682196339434e-06, | |
| "loss": 0.0078, | |
| "step": 379000 | |
| }, | |
| { | |
| "epoch": 157.86, | |
| "learning_rate": 4.2198835274542435e-06, | |
| "loss": 0.0078, | |
| "step": 379500 | |
| }, | |
| { | |
| "epoch": 158.0, | |
| "eval_cer": 0.243, | |
| "eval_gen_len": 13.7143, | |
| "eval_loss": 0.46363481879234314, | |
| "eval_runtime": 263.4944, | |
| "eval_samples_per_second": 12.167, | |
| "eval_steps_per_second": 3.044, | |
| "step": 379832 | |
| }, | |
| { | |
| "epoch": 158.07, | |
| "learning_rate": 4.199084858569052e-06, | |
| "loss": 0.0066, | |
| "step": 380000 | |
| }, | |
| { | |
| "epoch": 158.28, | |
| "learning_rate": 4.178286189683861e-06, | |
| "loss": 0.0073, | |
| "step": 380500 | |
| }, | |
| { | |
| "epoch": 158.49, | |
| "learning_rate": 4.15752911813644e-06, | |
| "loss": 0.0064, | |
| "step": 381000 | |
| }, | |
| { | |
| "epoch": 158.69, | |
| "learning_rate": 4.136730449251248e-06, | |
| "loss": 0.008, | |
| "step": 381500 | |
| }, | |
| { | |
| "epoch": 158.9, | |
| "learning_rate": 4.115931780366057e-06, | |
| "loss": 0.007, | |
| "step": 382000 | |
| }, | |
| { | |
| "epoch": 159.0, | |
| "eval_cer": 0.2413, | |
| "eval_gen_len": 13.6556, | |
| "eval_loss": 0.4499606192111969, | |
| "eval_runtime": 265.7001, | |
| "eval_samples_per_second": 12.066, | |
| "eval_steps_per_second": 3.018, | |
| "step": 382236 | |
| }, | |
| { | |
| "epoch": 159.11, | |
| "learning_rate": 4.095133111480866e-06, | |
| "loss": 0.0074, | |
| "step": 382500 | |
| }, | |
| { | |
| "epoch": 159.32, | |
| "learning_rate": 4.0743344425956745e-06, | |
| "loss": 0.0064, | |
| "step": 383000 | |
| }, | |
| { | |
| "epoch": 159.53, | |
| "learning_rate": 4.053535773710483e-06, | |
| "loss": 0.0068, | |
| "step": 383500 | |
| }, | |
| { | |
| "epoch": 159.73, | |
| "learning_rate": 4.032778702163062e-06, | |
| "loss": 0.0068, | |
| "step": 384000 | |
| }, | |
| { | |
| "epoch": 159.94, | |
| "learning_rate": 4.011980033277871e-06, | |
| "loss": 0.0079, | |
| "step": 384500 | |
| }, | |
| { | |
| "epoch": 160.0, | |
| "eval_cer": 0.241, | |
| "eval_gen_len": 13.622, | |
| "eval_loss": 0.4524941146373749, | |
| "eval_runtime": 263.8331, | |
| "eval_samples_per_second": 12.152, | |
| "eval_steps_per_second": 3.04, | |
| "step": 384640 | |
| }, | |
| { | |
| "epoch": 160.15, | |
| "learning_rate": 3.991181364392679e-06, | |
| "loss": 0.0076, | |
| "step": 385000 | |
| }, | |
| { | |
| "epoch": 160.36, | |
| "learning_rate": 3.9703826955074885e-06, | |
| "loss": 0.0062, | |
| "step": 385500 | |
| }, | |
| { | |
| "epoch": 160.57, | |
| "learning_rate": 3.949584026622296e-06, | |
| "loss": 0.0066, | |
| "step": 386000 | |
| }, | |
| { | |
| "epoch": 160.77, | |
| "learning_rate": 3.928785357737105e-06, | |
| "loss": 0.0067, | |
| "step": 386500 | |
| }, | |
| { | |
| "epoch": 160.98, | |
| "learning_rate": 3.908069883527454e-06, | |
| "loss": 0.0074, | |
| "step": 387000 | |
| }, | |
| { | |
| "epoch": 161.0, | |
| "eval_cer": 0.2439, | |
| "eval_gen_len": 13.7757, | |
| "eval_loss": 0.4422759711742401, | |
| "eval_runtime": 268.4253, | |
| "eval_samples_per_second": 11.944, | |
| "eval_steps_per_second": 2.988, | |
| "step": 387044 | |
| }, | |
| { | |
| "epoch": 161.19, | |
| "learning_rate": 3.887271214642263e-06, | |
| "loss": 0.008, | |
| "step": 387500 | |
| }, | |
| { | |
| "epoch": 161.4, | |
| "learning_rate": 3.866472545757072e-06, | |
| "loss": 0.0068, | |
| "step": 388000 | |
| }, | |
| { | |
| "epoch": 161.61, | |
| "learning_rate": 3.845673876871881e-06, | |
| "loss": 0.0066, | |
| "step": 388500 | |
| }, | |
| { | |
| "epoch": 161.81, | |
| "learning_rate": 3.824875207986689e-06, | |
| "loss": 0.0081, | |
| "step": 389000 | |
| }, | |
| { | |
| "epoch": 162.0, | |
| "eval_cer": 0.2433, | |
| "eval_gen_len": 13.7096, | |
| "eval_loss": 0.4686408042907715, | |
| "eval_runtime": 266.054, | |
| "eval_samples_per_second": 12.05, | |
| "eval_steps_per_second": 3.014, | |
| "step": 389448 | |
| }, | |
| { | |
| "epoch": 162.02, | |
| "learning_rate": 3.804118136439268e-06, | |
| "loss": 0.0073, | |
| "step": 389500 | |
| }, | |
| { | |
| "epoch": 162.23, | |
| "learning_rate": 3.7833194675540767e-06, | |
| "loss": 0.0057, | |
| "step": 390000 | |
| }, | |
| { | |
| "epoch": 162.44, | |
| "learning_rate": 3.7625207986688856e-06, | |
| "loss": 0.007, | |
| "step": 390500 | |
| }, | |
| { | |
| "epoch": 162.65, | |
| "learning_rate": 3.741722129783694e-06, | |
| "loss": 0.0074, | |
| "step": 391000 | |
| }, | |
| { | |
| "epoch": 162.85, | |
| "learning_rate": 3.7209234608985027e-06, | |
| "loss": 0.0067, | |
| "step": 391500 | |
| }, | |
| { | |
| "epoch": 163.0, | |
| "eval_cer": 0.2422, | |
| "eval_gen_len": 13.68, | |
| "eval_loss": 0.4667229950428009, | |
| "eval_runtime": 268.59, | |
| "eval_samples_per_second": 11.936, | |
| "eval_steps_per_second": 2.986, | |
| "step": 391852 | |
| }, | |
| { | |
| "epoch": 163.06, | |
| "learning_rate": 3.7001247920133116e-06, | |
| "loss": 0.0081, | |
| "step": 392000 | |
| }, | |
| { | |
| "epoch": 163.27, | |
| "learning_rate": 3.67932612312812e-06, | |
| "loss": 0.0067, | |
| "step": 392500 | |
| }, | |
| { | |
| "epoch": 163.48, | |
| "learning_rate": 3.6585274542429287e-06, | |
| "loss": 0.0064, | |
| "step": 393000 | |
| }, | |
| { | |
| "epoch": 163.69, | |
| "learning_rate": 3.6377287853577376e-06, | |
| "loss": 0.0073, | |
| "step": 393500 | |
| }, | |
| { | |
| "epoch": 163.89, | |
| "learning_rate": 3.6169717138103167e-06, | |
| "loss": 0.0074, | |
| "step": 394000 | |
| }, | |
| { | |
| "epoch": 164.0, | |
| "eval_cer": 0.2418, | |
| "eval_gen_len": 13.6482, | |
| "eval_loss": 0.4521939754486084, | |
| "eval_runtime": 273.0322, | |
| "eval_samples_per_second": 11.742, | |
| "eval_steps_per_second": 2.937, | |
| "step": 394256 | |
| }, | |
| { | |
| "epoch": 164.1, | |
| "learning_rate": 3.596173044925125e-06, | |
| "loss": 0.007, | |
| "step": 394500 | |
| }, | |
| { | |
| "epoch": 164.31, | |
| "learning_rate": 3.5754159733777038e-06, | |
| "loss": 0.0072, | |
| "step": 395000 | |
| }, | |
| { | |
| "epoch": 164.52, | |
| "learning_rate": 3.5546173044925127e-06, | |
| "loss": 0.0064, | |
| "step": 395500 | |
| }, | |
| { | |
| "epoch": 164.73, | |
| "learning_rate": 3.5338186356073212e-06, | |
| "loss": 0.0071, | |
| "step": 396000 | |
| }, | |
| { | |
| "epoch": 164.93, | |
| "learning_rate": 3.5130199667221298e-06, | |
| "loss": 0.0085, | |
| "step": 396500 | |
| }, | |
| { | |
| "epoch": 165.0, | |
| "eval_cer": 0.2414, | |
| "eval_gen_len": 13.6282, | |
| "eval_loss": 0.45710650086402893, | |
| "eval_runtime": 255.6061, | |
| "eval_samples_per_second": 12.543, | |
| "eval_steps_per_second": 3.138, | |
| "step": 396660 | |
| }, | |
| { | |
| "epoch": 165.14, | |
| "learning_rate": 3.4922212978369387e-06, | |
| "loss": 0.0063, | |
| "step": 397000 | |
| }, | |
| { | |
| "epoch": 165.35, | |
| "learning_rate": 3.4714226289517472e-06, | |
| "loss": 0.0068, | |
| "step": 397500 | |
| }, | |
| { | |
| "epoch": 165.56, | |
| "learning_rate": 3.4506239600665558e-06, | |
| "loss": 0.0072, | |
| "step": 398000 | |
| }, | |
| { | |
| "epoch": 165.77, | |
| "learning_rate": 3.4298668885191348e-06, | |
| "loss": 0.0065, | |
| "step": 398500 | |
| }, | |
| { | |
| "epoch": 165.97, | |
| "learning_rate": 3.4090682196339437e-06, | |
| "loss": 0.0074, | |
| "step": 399000 | |
| }, | |
| { | |
| "epoch": 166.0, | |
| "eval_cer": 0.241, | |
| "eval_gen_len": 13.6073, | |
| "eval_loss": 0.45994168519973755, | |
| "eval_runtime": 256.591, | |
| "eval_samples_per_second": 12.495, | |
| "eval_steps_per_second": 3.126, | |
| "step": 399064 | |
| }, | |
| { | |
| "epoch": 166.18, | |
| "learning_rate": 3.3882695507487522e-06, | |
| "loss": 0.0072, | |
| "step": 399500 | |
| }, | |
| { | |
| "epoch": 166.39, | |
| "learning_rate": 3.3674708818635608e-06, | |
| "loss": 0.0061, | |
| "step": 400000 | |
| }, | |
| { | |
| "epoch": 166.6, | |
| "learning_rate": 3.3466722129783697e-06, | |
| "loss": 0.0062, | |
| "step": 400500 | |
| }, | |
| { | |
| "epoch": 166.81, | |
| "learning_rate": 3.3258735440931782e-06, | |
| "loss": 0.0071, | |
| "step": 401000 | |
| }, | |
| { | |
| "epoch": 167.0, | |
| "eval_cer": 0.2427, | |
| "eval_gen_len": 13.5515, | |
| "eval_loss": 0.4614485502243042, | |
| "eval_runtime": 261.146, | |
| "eval_samples_per_second": 12.277, | |
| "eval_steps_per_second": 3.071, | |
| "step": 401468 | |
| }, | |
| { | |
| "epoch": 167.01, | |
| "learning_rate": 3.305074875207987e-06, | |
| "loss": 0.0076, | |
| "step": 401500 | |
| }, | |
| { | |
| "epoch": 167.22, | |
| "learning_rate": 3.2842762063227957e-06, | |
| "loss": 0.0068, | |
| "step": 402000 | |
| }, | |
| { | |
| "epoch": 167.43, | |
| "learning_rate": 3.2634775374376042e-06, | |
| "loss": 0.0075, | |
| "step": 402500 | |
| }, | |
| { | |
| "epoch": 167.64, | |
| "learning_rate": 3.2427204658901832e-06, | |
| "loss": 0.007, | |
| "step": 403000 | |
| }, | |
| { | |
| "epoch": 167.85, | |
| "learning_rate": 3.221921797004992e-06, | |
| "loss": 0.0054, | |
| "step": 403500 | |
| }, | |
| { | |
| "epoch": 168.0, | |
| "eval_cer": 0.2471, | |
| "eval_gen_len": 13.8865, | |
| "eval_loss": 0.4741056263446808, | |
| "eval_runtime": 274.0865, | |
| "eval_samples_per_second": 11.697, | |
| "eval_steps_per_second": 2.926, | |
| "step": 403872 | |
| }, | |
| { | |
| "epoch": 168.05, | |
| "learning_rate": 3.2011231281198007e-06, | |
| "loss": 0.0066, | |
| "step": 404000 | |
| }, | |
| { | |
| "epoch": 168.26, | |
| "learning_rate": 3.1803244592346092e-06, | |
| "loss": 0.0058, | |
| "step": 404500 | |
| }, | |
| { | |
| "epoch": 168.47, | |
| "learning_rate": 3.1595673876871887e-06, | |
| "loss": 0.0074, | |
| "step": 405000 | |
| }, | |
| { | |
| "epoch": 168.68, | |
| "learning_rate": 3.138768718801997e-06, | |
| "loss": 0.007, | |
| "step": 405500 | |
| }, | |
| { | |
| "epoch": 168.89, | |
| "learning_rate": 3.1179700499168057e-06, | |
| "loss": 0.0076, | |
| "step": 406000 | |
| }, | |
| { | |
| "epoch": 169.0, | |
| "eval_cer": 0.2432, | |
| "eval_gen_len": 13.6996, | |
| "eval_loss": 0.46461164951324463, | |
| "eval_runtime": 266.1722, | |
| "eval_samples_per_second": 12.045, | |
| "eval_steps_per_second": 3.013, | |
| "step": 406276 | |
| }, | |
| { | |
| "epoch": 169.09, | |
| "learning_rate": 3.097171381031614e-06, | |
| "loss": 0.0065, | |
| "step": 406500 | |
| }, | |
| { | |
| "epoch": 169.3, | |
| "learning_rate": 3.0763727121464228e-06, | |
| "loss": 0.0061, | |
| "step": 407000 | |
| }, | |
| { | |
| "epoch": 169.51, | |
| "learning_rate": 3.0555740432612313e-06, | |
| "loss": 0.0067, | |
| "step": 407500 | |
| }, | |
| { | |
| "epoch": 169.72, | |
| "learning_rate": 3.03477537437604e-06, | |
| "loss": 0.0075, | |
| "step": 408000 | |
| }, | |
| { | |
| "epoch": 169.93, | |
| "learning_rate": 3.0140183028286193e-06, | |
| "loss": 0.0064, | |
| "step": 408500 | |
| }, | |
| { | |
| "epoch": 170.0, | |
| "eval_cer": 0.2407, | |
| "eval_gen_len": 13.6525, | |
| "eval_loss": 0.45978671312332153, | |
| "eval_runtime": 264.0025, | |
| "eval_samples_per_second": 12.144, | |
| "eval_steps_per_second": 3.038, | |
| "step": 408680 | |
| }, | |
| { | |
| "epoch": 170.13, | |
| "learning_rate": 2.993219633943428e-06, | |
| "loss": 0.0058, | |
| "step": 409000 | |
| }, | |
| { | |
| "epoch": 170.34, | |
| "learning_rate": 2.9724209650582363e-06, | |
| "loss": 0.0072, | |
| "step": 409500 | |
| }, | |
| { | |
| "epoch": 170.55, | |
| "learning_rate": 2.9516222961730453e-06, | |
| "loss": 0.0073, | |
| "step": 410000 | |
| }, | |
| { | |
| "epoch": 170.76, | |
| "learning_rate": 2.9308236272878538e-06, | |
| "loss": 0.0067, | |
| "step": 410500 | |
| }, | |
| { | |
| "epoch": 170.97, | |
| "learning_rate": 2.9100249584026623e-06, | |
| "loss": 0.0066, | |
| "step": 411000 | |
| }, | |
| { | |
| "epoch": 171.0, | |
| "eval_cer": 0.2463, | |
| "eval_gen_len": 13.8612, | |
| "eval_loss": 0.4791421890258789, | |
| "eval_runtime": 266.2729, | |
| "eval_samples_per_second": 12.04, | |
| "eval_steps_per_second": 3.012, | |
| "step": 411084 | |
| }, | |
| { | |
| "epoch": 171.17, | |
| "learning_rate": 2.8892262895174713e-06, | |
| "loss": 0.0066, | |
| "step": 411500 | |
| }, | |
| { | |
| "epoch": 171.38, | |
| "learning_rate": 2.8684276206322798e-06, | |
| "loss": 0.0055, | |
| "step": 412000 | |
| }, | |
| { | |
| "epoch": 171.59, | |
| "learning_rate": 2.847670549084859e-06, | |
| "loss": 0.0072, | |
| "step": 412500 | |
| }, | |
| { | |
| "epoch": 171.8, | |
| "learning_rate": 2.8268718801996673e-06, | |
| "loss": 0.0067, | |
| "step": 413000 | |
| }, | |
| { | |
| "epoch": 172.0, | |
| "eval_cer": 0.2408, | |
| "eval_gen_len": 13.6622, | |
| "eval_loss": 0.4588078260421753, | |
| "eval_runtime": 266.7256, | |
| "eval_samples_per_second": 12.02, | |
| "eval_steps_per_second": 3.007, | |
| "step": 413488 | |
| }, | |
| { | |
| "epoch": 172.0, | |
| "learning_rate": 2.8060732113144763e-06, | |
| "loss": 0.0074, | |
| "step": 413500 | |
| }, | |
| { | |
| "epoch": 172.21, | |
| "learning_rate": 2.7852745424292848e-06, | |
| "loss": 0.0066, | |
| "step": 414000 | |
| }, | |
| { | |
| "epoch": 172.42, | |
| "learning_rate": 2.7644758735440937e-06, | |
| "loss": 0.0058, | |
| "step": 414500 | |
| }, | |
| { | |
| "epoch": 172.63, | |
| "learning_rate": 2.7436772046589023e-06, | |
| "loss": 0.0067, | |
| "step": 415000 | |
| }, | |
| { | |
| "epoch": 172.84, | |
| "learning_rate": 2.7228785357737108e-06, | |
| "loss": 0.0074, | |
| "step": 415500 | |
| }, | |
| { | |
| "epoch": 173.0, | |
| "eval_cer": 0.2411, | |
| "eval_gen_len": 13.7199, | |
| "eval_loss": 0.45347917079925537, | |
| "eval_runtime": 266.7945, | |
| "eval_samples_per_second": 12.017, | |
| "eval_steps_per_second": 3.006, | |
| "step": 415892 | |
| }, | |
| { | |
| "epoch": 173.04, | |
| "learning_rate": 2.70212146422629e-06, | |
| "loss": 0.0065, | |
| "step": 416000 | |
| }, | |
| { | |
| "epoch": 173.25, | |
| "learning_rate": 2.6813227953410987e-06, | |
| "loss": 0.0066, | |
| "step": 416500 | |
| }, | |
| { | |
| "epoch": 173.46, | |
| "learning_rate": 2.6605241264559073e-06, | |
| "loss": 0.006, | |
| "step": 417000 | |
| }, | |
| { | |
| "epoch": 173.67, | |
| "learning_rate": 2.639725457570716e-06, | |
| "loss": 0.0065, | |
| "step": 417500 | |
| }, | |
| { | |
| "epoch": 173.88, | |
| "learning_rate": 2.618926788685524e-06, | |
| "loss": 0.0073, | |
| "step": 418000 | |
| }, | |
| { | |
| "epoch": 174.0, | |
| "eval_cer": 0.2413, | |
| "eval_gen_len": 13.7389, | |
| "eval_loss": 0.4470750093460083, | |
| "eval_runtime": 261.7235, | |
| "eval_samples_per_second": 12.25, | |
| "eval_steps_per_second": 3.064, | |
| "step": 418296 | |
| }, | |
| { | |
| "epoch": 174.08, | |
| "learning_rate": 2.598128119800333e-06, | |
| "loss": 0.007, | |
| "step": 418500 | |
| }, | |
| { | |
| "epoch": 174.29, | |
| "learning_rate": 2.5773294509151414e-06, | |
| "loss": 0.0068, | |
| "step": 419000 | |
| }, | |
| { | |
| "epoch": 174.5, | |
| "learning_rate": 2.55653078202995e-06, | |
| "loss": 0.0073, | |
| "step": 419500 | |
| }, | |
| { | |
| "epoch": 174.71, | |
| "learning_rate": 2.5358153078203e-06, | |
| "loss": 0.0057, | |
| "step": 420000 | |
| }, | |
| { | |
| "epoch": 174.92, | |
| "learning_rate": 2.5150166389351083e-06, | |
| "loss": 0.0066, | |
| "step": 420500 | |
| }, | |
| { | |
| "epoch": 175.0, | |
| "eval_cer": 0.2421, | |
| "eval_gen_len": 13.7676, | |
| "eval_loss": 0.46230319142341614, | |
| "eval_runtime": 274.1203, | |
| "eval_samples_per_second": 11.696, | |
| "eval_steps_per_second": 2.926, | |
| "step": 420700 | |
| }, | |
| { | |
| "epoch": 175.12, | |
| "learning_rate": 2.494217970049917e-06, | |
| "loss": 0.0072, | |
| "step": 421000 | |
| }, | |
| { | |
| "epoch": 175.33, | |
| "learning_rate": 2.473419301164726e-06, | |
| "loss": 0.006, | |
| "step": 421500 | |
| }, | |
| { | |
| "epoch": 175.54, | |
| "learning_rate": 2.4526206322795343e-06, | |
| "loss": 0.0068, | |
| "step": 422000 | |
| }, | |
| { | |
| "epoch": 175.75, | |
| "learning_rate": 2.4318635607321134e-06, | |
| "loss": 0.0068, | |
| "step": 422500 | |
| }, | |
| { | |
| "epoch": 175.96, | |
| "learning_rate": 2.411064891846922e-06, | |
| "loss": 0.0067, | |
| "step": 423000 | |
| }, | |
| { | |
| "epoch": 176.0, | |
| "eval_cer": 0.2427, | |
| "eval_gen_len": 13.7302, | |
| "eval_loss": 0.46741247177124023, | |
| "eval_runtime": 271.0852, | |
| "eval_samples_per_second": 11.827, | |
| "eval_steps_per_second": 2.958, | |
| "step": 423104 | |
| }, | |
| { | |
| "epoch": 176.16, | |
| "learning_rate": 2.390266222961731e-06, | |
| "loss": 0.0065, | |
| "step": 423500 | |
| }, | |
| { | |
| "epoch": 176.37, | |
| "learning_rate": 2.3694675540765393e-06, | |
| "loss": 0.0062, | |
| "step": 424000 | |
| }, | |
| { | |
| "epoch": 176.58, | |
| "learning_rate": 2.348668885191348e-06, | |
| "loss": 0.0057, | |
| "step": 424500 | |
| }, | |
| { | |
| "epoch": 176.79, | |
| "learning_rate": 2.327870216306157e-06, | |
| "loss": 0.007, | |
| "step": 425000 | |
| }, | |
| { | |
| "epoch": 177.0, | |
| "learning_rate": 2.307071547420965e-06, | |
| "loss": 0.0077, | |
| "step": 425500 | |
| }, | |
| { | |
| "epoch": 177.0, | |
| "eval_cer": 0.2399, | |
| "eval_gen_len": 13.5359, | |
| "eval_loss": 0.45084038376808167, | |
| "eval_runtime": 262.6698, | |
| "eval_samples_per_second": 12.205, | |
| "eval_steps_per_second": 3.053, | |
| "step": 425508 | |
| }, | |
| { | |
| "epoch": 177.2, | |
| "learning_rate": 2.286272878535774e-06, | |
| "loss": 0.0059, | |
| "step": 426000 | |
| }, | |
| { | |
| "epoch": 177.41, | |
| "learning_rate": 2.2654742096505824e-06, | |
| "loss": 0.0067, | |
| "step": 426500 | |
| }, | |
| { | |
| "epoch": 177.62, | |
| "learning_rate": 2.2447171381031614e-06, | |
| "loss": 0.006, | |
| "step": 427000 | |
| }, | |
| { | |
| "epoch": 177.83, | |
| "learning_rate": 2.2239184692179704e-06, | |
| "loss": 0.0066, | |
| "step": 427500 | |
| }, | |
| { | |
| "epoch": 178.0, | |
| "eval_cer": 0.2408, | |
| "eval_gen_len": 13.7302, | |
| "eval_loss": 0.4653932452201843, | |
| "eval_runtime": 268.3858, | |
| "eval_samples_per_second": 11.945, | |
| "eval_steps_per_second": 2.988, | |
| "step": 427912 | |
| }, | |
| { | |
| "epoch": 178.04, | |
| "learning_rate": 2.203119800332779e-06, | |
| "loss": 0.0064, | |
| "step": 428000 | |
| }, | |
| { | |
| "epoch": 178.24, | |
| "learning_rate": 2.1823211314475874e-06, | |
| "loss": 0.0055, | |
| "step": 428500 | |
| }, | |
| { | |
| "epoch": 178.45, | |
| "learning_rate": 2.161564059900167e-06, | |
| "loss": 0.0068, | |
| "step": 429000 | |
| }, | |
| { | |
| "epoch": 178.66, | |
| "learning_rate": 2.1407653910149754e-06, | |
| "loss": 0.0068, | |
| "step": 429500 | |
| }, | |
| { | |
| "epoch": 178.87, | |
| "learning_rate": 2.119966722129784e-06, | |
| "loss": 0.0067, | |
| "step": 430000 | |
| }, | |
| { | |
| "epoch": 179.0, | |
| "eval_cer": 0.2408, | |
| "eval_gen_len": 13.6925, | |
| "eval_loss": 0.462500661611557, | |
| "eval_runtime": 268.4086, | |
| "eval_samples_per_second": 11.944, | |
| "eval_steps_per_second": 2.988, | |
| "step": 430316 | |
| }, | |
| { | |
| "epoch": 179.08, | |
| "learning_rate": 2.0991680532445924e-06, | |
| "loss": 0.0067, | |
| "step": 430500 | |
| }, | |
| { | |
| "epoch": 179.28, | |
| "learning_rate": 2.078369384359401e-06, | |
| "loss": 0.0059, | |
| "step": 431000 | |
| }, | |
| { | |
| "epoch": 179.49, | |
| "learning_rate": 2.05761231281198e-06, | |
| "loss": 0.0064, | |
| "step": 431500 | |
| }, | |
| { | |
| "epoch": 179.7, | |
| "learning_rate": 2.036813643926789e-06, | |
| "loss": 0.0065, | |
| "step": 432000 | |
| }, | |
| { | |
| "epoch": 179.91, | |
| "learning_rate": 2.0160149750415974e-06, | |
| "loss": 0.0072, | |
| "step": 432500 | |
| }, | |
| { | |
| "epoch": 180.0, | |
| "eval_cer": 0.242, | |
| "eval_gen_len": 13.7545, | |
| "eval_loss": 0.46416959166526794, | |
| "eval_runtime": 267.9605, | |
| "eval_samples_per_second": 11.964, | |
| "eval_steps_per_second": 2.993, | |
| "step": 432720 | |
| }, | |
| { | |
| "epoch": 180.12, | |
| "learning_rate": 1.995216306156406e-06, | |
| "loss": 0.0064, | |
| "step": 433000 | |
| }, | |
| { | |
| "epoch": 180.32, | |
| "learning_rate": 1.9744592346089854e-06, | |
| "loss": 0.0058, | |
| "step": 433500 | |
| }, | |
| { | |
| "epoch": 180.53, | |
| "learning_rate": 1.953660565723794e-06, | |
| "loss": 0.0063, | |
| "step": 434000 | |
| }, | |
| { | |
| "epoch": 180.74, | |
| "learning_rate": 1.9328618968386024e-06, | |
| "loss": 0.0065, | |
| "step": 434500 | |
| }, | |
| { | |
| "epoch": 180.95, | |
| "learning_rate": 1.9120632279534114e-06, | |
| "loss": 0.0066, | |
| "step": 435000 | |
| }, | |
| { | |
| "epoch": 181.0, | |
| "eval_cer": 0.2401, | |
| "eval_gen_len": 13.6538, | |
| "eval_loss": 0.4692617356777191, | |
| "eval_runtime": 260.4012, | |
| "eval_samples_per_second": 12.312, | |
| "eval_steps_per_second": 3.08, | |
| "step": 435124 | |
| }, | |
| { | |
| "epoch": 181.16, | |
| "learning_rate": 1.8912645590682197e-06, | |
| "loss": 0.0076, | |
| "step": 435500 | |
| }, | |
| { | |
| "epoch": 181.36, | |
| "learning_rate": 1.8705074875207987e-06, | |
| "loss": 0.0059, | |
| "step": 436000 | |
| }, | |
| { | |
| "epoch": 181.57, | |
| "learning_rate": 1.8497088186356074e-06, | |
| "loss": 0.0069, | |
| "step": 436500 | |
| }, | |
| { | |
| "epoch": 181.78, | |
| "learning_rate": 1.8289101497504162e-06, | |
| "loss": 0.0061, | |
| "step": 437000 | |
| }, | |
| { | |
| "epoch": 181.99, | |
| "learning_rate": 1.8081114808652247e-06, | |
| "loss": 0.0064, | |
| "step": 437500 | |
| }, | |
| { | |
| "epoch": 182.0, | |
| "eval_cer": 0.2413, | |
| "eval_gen_len": 13.7302, | |
| "eval_loss": 0.4686383605003357, | |
| "eval_runtime": 260.1458, | |
| "eval_samples_per_second": 12.324, | |
| "eval_steps_per_second": 3.083, | |
| "step": 437528 | |
| }, | |
| { | |
| "epoch": 182.2, | |
| "learning_rate": 1.787354409317804e-06, | |
| "loss": 0.0062, | |
| "step": 438000 | |
| }, | |
| { | |
| "epoch": 182.4, | |
| "learning_rate": 1.7665557404326125e-06, | |
| "loss": 0.0066, | |
| "step": 438500 | |
| }, | |
| { | |
| "epoch": 182.61, | |
| "learning_rate": 1.7457570715474212e-06, | |
| "loss": 0.0066, | |
| "step": 439000 | |
| }, | |
| { | |
| "epoch": 182.82, | |
| "learning_rate": 1.725e-06, | |
| "loss": 0.0067, | |
| "step": 439500 | |
| }, | |
| { | |
| "epoch": 183.0, | |
| "eval_cer": 0.2415, | |
| "eval_gen_len": 13.7371, | |
| "eval_loss": 0.46899163722991943, | |
| "eval_runtime": 270.7897, | |
| "eval_samples_per_second": 11.839, | |
| "eval_steps_per_second": 2.962, | |
| "step": 439932 | |
| }, | |
| { | |
| "epoch": 183.03, | |
| "learning_rate": 1.7042013311148087e-06, | |
| "loss": 0.0059, | |
| "step": 440000 | |
| }, | |
| { | |
| "epoch": 183.24, | |
| "learning_rate": 1.6834026622296173e-06, | |
| "loss": 0.0057, | |
| "step": 440500 | |
| }, | |
| { | |
| "epoch": 183.44, | |
| "learning_rate": 1.662603993344426e-06, | |
| "loss": 0.0062, | |
| "step": 441000 | |
| }, | |
| { | |
| "epoch": 183.65, | |
| "learning_rate": 1.6418053244592347e-06, | |
| "loss": 0.0061, | |
| "step": 441500 | |
| }, | |
| { | |
| "epoch": 183.86, | |
| "learning_rate": 1.6210066555740433e-06, | |
| "loss": 0.0067, | |
| "step": 442000 | |
| }, | |
| { | |
| "epoch": 184.0, | |
| "eval_cer": 0.2392, | |
| "eval_gen_len": 13.7442, | |
| "eval_loss": 0.47148939967155457, | |
| "eval_runtime": 271.9757, | |
| "eval_samples_per_second": 11.788, | |
| "eval_steps_per_second": 2.949, | |
| "step": 442336 | |
| }, | |
| { | |
| "epoch": 184.07, | |
| "learning_rate": 1.600207986688852e-06, | |
| "loss": 0.0066, | |
| "step": 442500 | |
| }, | |
| { | |
| "epoch": 184.28, | |
| "learning_rate": 1.5794093178036607e-06, | |
| "loss": 0.0063, | |
| "step": 443000 | |
| }, | |
| { | |
| "epoch": 184.48, | |
| "learning_rate": 1.5586106489184695e-06, | |
| "loss": 0.0063, | |
| "step": 443500 | |
| }, | |
| { | |
| "epoch": 184.69, | |
| "learning_rate": 1.5378535773710485e-06, | |
| "loss": 0.0063, | |
| "step": 444000 | |
| }, | |
| { | |
| "epoch": 184.9, | |
| "learning_rate": 1.5170965058236273e-06, | |
| "loss": 0.0062, | |
| "step": 444500 | |
| }, | |
| { | |
| "epoch": 185.0, | |
| "eval_cer": 0.2395, | |
| "eval_gen_len": 13.6572, | |
| "eval_loss": 0.46137315034866333, | |
| "eval_runtime": 269.0871, | |
| "eval_samples_per_second": 11.914, | |
| "eval_steps_per_second": 2.98, | |
| "step": 444740 | |
| }, | |
| { | |
| "epoch": 185.11, | |
| "learning_rate": 1.496297836938436e-06, | |
| "loss": 0.0066, | |
| "step": 445000 | |
| }, | |
| { | |
| "epoch": 185.32, | |
| "learning_rate": 1.4754991680532445e-06, | |
| "loss": 0.006, | |
| "step": 445500 | |
| }, | |
| { | |
| "epoch": 185.52, | |
| "learning_rate": 1.4547004991680533e-06, | |
| "loss": 0.0053, | |
| "step": 446000 | |
| }, | |
| { | |
| "epoch": 185.73, | |
| "learning_rate": 1.433901830282862e-06, | |
| "loss": 0.0075, | |
| "step": 446500 | |
| }, | |
| { | |
| "epoch": 185.94, | |
| "learning_rate": 1.4131031613976705e-06, | |
| "loss": 0.0068, | |
| "step": 447000 | |
| }, | |
| { | |
| "epoch": 186.0, | |
| "eval_cer": 0.2396, | |
| "eval_gen_len": 13.6185, | |
| "eval_loss": 0.46076661348342896, | |
| "eval_runtime": 256.715, | |
| "eval_samples_per_second": 12.489, | |
| "eval_steps_per_second": 3.124, | |
| "step": 447144 | |
| }, | |
| { | |
| "epoch": 186.15, | |
| "learning_rate": 1.3923044925124793e-06, | |
| "loss": 0.0062, | |
| "step": 447500 | |
| }, | |
| { | |
| "epoch": 186.36, | |
| "learning_rate": 1.371505823627288e-06, | |
| "loss": 0.0058, | |
| "step": 448000 | |
| }, | |
| { | |
| "epoch": 186.56, | |
| "learning_rate": 1.3507071547420965e-06, | |
| "loss": 0.0065, | |
| "step": 448500 | |
| }, | |
| { | |
| "epoch": 186.77, | |
| "learning_rate": 1.3299500831946758e-06, | |
| "loss": 0.0065, | |
| "step": 449000 | |
| }, | |
| { | |
| "epoch": 186.98, | |
| "learning_rate": 1.3091514143094845e-06, | |
| "loss": 0.0064, | |
| "step": 449500 | |
| }, | |
| { | |
| "epoch": 187.0, | |
| "eval_cer": 0.2391, | |
| "eval_gen_len": 13.7558, | |
| "eval_loss": 0.47285133600234985, | |
| "eval_runtime": 259.8275, | |
| "eval_samples_per_second": 12.339, | |
| "eval_steps_per_second": 3.087, | |
| "step": 449548 | |
| }, | |
| { | |
| "epoch": 187.19, | |
| "learning_rate": 1.288352745424293e-06, | |
| "loss": 0.0054, | |
| "step": 450000 | |
| }, | |
| { | |
| "epoch": 187.4, | |
| "learning_rate": 1.2675540765391017e-06, | |
| "loss": 0.0064, | |
| "step": 450500 | |
| }, | |
| { | |
| "epoch": 187.6, | |
| "learning_rate": 1.2467554076539103e-06, | |
| "loss": 0.0059, | |
| "step": 451000 | |
| }, | |
| { | |
| "epoch": 187.81, | |
| "learning_rate": 1.225956738768719e-06, | |
| "loss": 0.0059, | |
| "step": 451500 | |
| }, | |
| { | |
| "epoch": 188.0, | |
| "eval_cer": 0.2397, | |
| "eval_gen_len": 13.6946, | |
| "eval_loss": 0.47259289026260376, | |
| "eval_runtime": 273.3002, | |
| "eval_samples_per_second": 11.731, | |
| "eval_steps_per_second": 2.935, | |
| "step": 451952 | |
| }, | |
| { | |
| "epoch": 188.02, | |
| "learning_rate": 1.2051996672212978e-06, | |
| "loss": 0.0073, | |
| "step": 452000 | |
| }, | |
| { | |
| "epoch": 188.23, | |
| "learning_rate": 1.1844009983361065e-06, | |
| "loss": 0.0065, | |
| "step": 452500 | |
| }, | |
| { | |
| "epoch": 188.44, | |
| "learning_rate": 1.1636439267886856e-06, | |
| "loss": 0.0069, | |
| "step": 453000 | |
| }, | |
| { | |
| "epoch": 188.64, | |
| "learning_rate": 1.1428452579034943e-06, | |
| "loss": 0.0062, | |
| "step": 453500 | |
| }, | |
| { | |
| "epoch": 188.85, | |
| "learning_rate": 1.1220881863560733e-06, | |
| "loss": 0.0052, | |
| "step": 454000 | |
| }, | |
| { | |
| "epoch": 189.0, | |
| "eval_cer": 0.239, | |
| "eval_gen_len": 13.6185, | |
| "eval_loss": 0.4666392505168915, | |
| "eval_runtime": 268.978, | |
| "eval_samples_per_second": 11.919, | |
| "eval_steps_per_second": 2.982, | |
| "step": 454356 | |
| }, | |
| { | |
| "epoch": 189.06, | |
| "learning_rate": 1.1012895174708818e-06, | |
| "loss": 0.0061, | |
| "step": 454500 | |
| }, | |
| { | |
| "epoch": 189.27, | |
| "learning_rate": 1.0804908485856906e-06, | |
| "loss": 0.0065, | |
| "step": 455000 | |
| }, | |
| { | |
| "epoch": 189.48, | |
| "learning_rate": 1.0596921797004993e-06, | |
| "loss": 0.0051, | |
| "step": 455500 | |
| }, | |
| { | |
| "epoch": 189.68, | |
| "learning_rate": 1.038893510815308e-06, | |
| "loss": 0.0066, | |
| "step": 456000 | |
| }, | |
| { | |
| "epoch": 189.89, | |
| "learning_rate": 1.0180948419301166e-06, | |
| "loss": 0.0066, | |
| "step": 456500 | |
| }, | |
| { | |
| "epoch": 190.0, | |
| "eval_cer": 0.2381, | |
| "eval_gen_len": 13.6825, | |
| "eval_loss": 0.4684942066669464, | |
| "eval_runtime": 269.3222, | |
| "eval_samples_per_second": 11.904, | |
| "eval_steps_per_second": 2.978, | |
| "step": 456760 | |
| }, | |
| { | |
| "epoch": 190.1, | |
| "learning_rate": 9.97296173044925e-07, | |
| "loss": 0.0061, | |
| "step": 457000 | |
| }, | |
| { | |
| "epoch": 190.31, | |
| "learning_rate": 9.765391014975043e-07, | |
| "loss": 0.0058, | |
| "step": 457500 | |
| }, | |
| { | |
| "epoch": 190.52, | |
| "learning_rate": 9.557404326123129e-07, | |
| "loss": 0.0058, | |
| "step": 458000 | |
| }, | |
| { | |
| "epoch": 190.72, | |
| "learning_rate": 9.349417637271216e-07, | |
| "loss": 0.0065, | |
| "step": 458500 | |
| }, | |
| { | |
| "epoch": 190.93, | |
| "learning_rate": 9.141430948419302e-07, | |
| "loss": 0.0061, | |
| "step": 459000 | |
| }, | |
| { | |
| "epoch": 191.0, | |
| "eval_cer": 0.238, | |
| "eval_gen_len": 13.6538, | |
| "eval_loss": 0.46481847763061523, | |
| "eval_runtime": 260.0695, | |
| "eval_samples_per_second": 12.327, | |
| "eval_steps_per_second": 3.084, | |
| "step": 459164 | |
| }, | |
| { | |
| "epoch": 191.14, | |
| "learning_rate": 8.933444259567387e-07, | |
| "loss": 0.006, | |
| "step": 459500 | |
| }, | |
| { | |
| "epoch": 191.35, | |
| "learning_rate": 8.725457570715475e-07, | |
| "loss": 0.0055, | |
| "step": 460000 | |
| }, | |
| { | |
| "epoch": 191.56, | |
| "learning_rate": 8.517470881863561e-07, | |
| "loss": 0.0058, | |
| "step": 460500 | |
| }, | |
| { | |
| "epoch": 191.76, | |
| "learning_rate": 8.309900166389352e-07, | |
| "loss": 0.0066, | |
| "step": 461000 | |
| }, | |
| { | |
| "epoch": 191.97, | |
| "learning_rate": 8.101913477537439e-07, | |
| "loss": 0.0063, | |
| "step": 461500 | |
| }, | |
| { | |
| "epoch": 192.0, | |
| "eval_cer": 0.2386, | |
| "eval_gen_len": 13.6301, | |
| "eval_loss": 0.46835771203041077, | |
| "eval_runtime": 258.927, | |
| "eval_samples_per_second": 12.382, | |
| "eval_steps_per_second": 3.097, | |
| "step": 461568 | |
| }, | |
| { | |
| "epoch": 192.18, | |
| "learning_rate": 7.893926788685524e-07, | |
| "loss": 0.0063, | |
| "step": 462000 | |
| }, | |
| { | |
| "epoch": 192.39, | |
| "learning_rate": 7.685940099833611e-07, | |
| "loss": 0.0059, | |
| "step": 462500 | |
| }, | |
| { | |
| "epoch": 192.6, | |
| "learning_rate": 7.477953410981697e-07, | |
| "loss": 0.0058, | |
| "step": 463000 | |
| }, | |
| { | |
| "epoch": 192.8, | |
| "learning_rate": 7.269966722129785e-07, | |
| "loss": 0.0064, | |
| "step": 463500 | |
| }, | |
| { | |
| "epoch": 193.0, | |
| "eval_cer": 0.2377, | |
| "eval_gen_len": 13.6687, | |
| "eval_loss": 0.4715929627418518, | |
| "eval_runtime": 259.4667, | |
| "eval_samples_per_second": 12.356, | |
| "eval_steps_per_second": 3.091, | |
| "step": 463972 | |
| }, | |
| { | |
| "epoch": 193.01, | |
| "learning_rate": 7.061980033277871e-07, | |
| "loss": 0.0051, | |
| "step": 464000 | |
| }, | |
| { | |
| "epoch": 193.22, | |
| "learning_rate": 6.853993344425957e-07, | |
| "loss": 0.0054, | |
| "step": 464500 | |
| } | |
| ], | |
| "max_steps": 480800, | |
| "num_train_epochs": 200, | |
| "total_flos": 2079248374038528.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |