Publish model from DewiBrynJones/whisper-large-v2-ft-cy-2603 revision 2603-20260310
d3f656b verified | { | |
| "best_global_step": 7000, | |
| "best_metric": 0.36715638637542725, | |
| "best_model_checkpoint": "/models/hfhub/DewiBrynJones/whisper-large-v2-ft-cy-2603/checkpoint-7000", | |
| "epoch": 3.3444816053511706, | |
| "eval_steps": 500, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.020903010033444816, | |
| "grad_norm": 3.98636531829834, | |
| "learning_rate": 9.800000000000001e-07, | |
| "loss": 1.4189, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04180602006688963, | |
| "grad_norm": 4.2314863204956055, | |
| "learning_rate": 1.98e-06, | |
| "loss": 0.8765, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06270903010033445, | |
| "grad_norm": 4.152566432952881, | |
| "learning_rate": 2.9800000000000003e-06, | |
| "loss": 0.7414, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.08361204013377926, | |
| "grad_norm": 3.3204166889190674, | |
| "learning_rate": 3.980000000000001e-06, | |
| "loss": 0.7041, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.10451505016722408, | |
| "grad_norm": 3.3151895999908447, | |
| "learning_rate": 4.980000000000001e-06, | |
| "loss": 0.6798, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1254180602006689, | |
| "grad_norm": 3.4774937629699707, | |
| "learning_rate": 5.98e-06, | |
| "loss": 0.6464, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.14632107023411373, | |
| "grad_norm": 3.0979840755462646, | |
| "learning_rate": 6.98e-06, | |
| "loss": 0.6288, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.16722408026755853, | |
| "grad_norm": 4.075951099395752, | |
| "learning_rate": 7.980000000000002e-06, | |
| "loss": 0.614, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.18812709030100336, | |
| "grad_norm": 2.7047572135925293, | |
| "learning_rate": 8.98e-06, | |
| "loss": 0.6035, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.20903010033444816, | |
| "grad_norm": 3.290921688079834, | |
| "learning_rate": 9.980000000000001e-06, | |
| "loss": 0.5905, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.20903010033444816, | |
| "eval_loss": 0.6038311719894409, | |
| "eval_runtime": 787.6035, | |
| "eval_samples_per_second": 4.945, | |
| "eval_steps_per_second": 0.155, | |
| "eval_wer": 0.4385438474063736, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.229933110367893, | |
| "grad_norm": 2.9369146823883057, | |
| "learning_rate": 9.998946840521706e-06, | |
| "loss": 0.5775, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.2508361204013378, | |
| "grad_norm": 3.506248950958252, | |
| "learning_rate": 9.9957014163915e-06, | |
| "loss": 0.5655, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.2717391304347826, | |
| "grad_norm": 2.5317389965057373, | |
| "learning_rate": 9.990264712510586e-06, | |
| "loss": 0.5429, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.29264214046822745, | |
| "grad_norm": 2.3587021827697754, | |
| "learning_rate": 9.982639113596978e-06, | |
| "loss": 0.5334, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.31354515050167225, | |
| "grad_norm": 2.6179604530334473, | |
| "learning_rate": 9.972827964490382e-06, | |
| "loss": 0.522, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.33444816053511706, | |
| "grad_norm": 2.407919406890869, | |
| "learning_rate": 9.960835568685058e-06, | |
| "loss": 0.5183, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.35535117056856186, | |
| "grad_norm": 2.5492148399353027, | |
| "learning_rate": 9.946667186442162e-06, | |
| "loss": 0.5157, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.3762541806020067, | |
| "grad_norm": 2.4506359100341797, | |
| "learning_rate": 9.930329032482412e-06, | |
| "loss": 0.5174, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.3971571906354515, | |
| "grad_norm": 2.404939889907837, | |
| "learning_rate": 9.911828273260119e-06, | |
| "loss": 0.4989, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.4180602006688963, | |
| "grad_norm": 2.676882028579712, | |
| "learning_rate": 9.891173023819731e-06, | |
| "loss": 0.4956, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4180602006688963, | |
| "eval_loss": 0.5025212168693542, | |
| "eval_runtime": 850.503, | |
| "eval_samples_per_second": 4.58, | |
| "eval_steps_per_second": 0.143, | |
| "eval_wer": 0.42074901139799953, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4389632107023411, | |
| "grad_norm": 2.4262278079986572, | |
| "learning_rate": 9.868372344236314e-06, | |
| "loss": 0.4955, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.459866220735786, | |
| "grad_norm": 2.3477671146392822, | |
| "learning_rate": 9.843436235641506e-06, | |
| "loss": 0.4717, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.4807692307692308, | |
| "grad_norm": 3.400567054748535, | |
| "learning_rate": 9.816375635836683e-06, | |
| "loss": 0.4668, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.5016722408026756, | |
| "grad_norm": 2.5671420097351074, | |
| "learning_rate": 9.787202414495275e-06, | |
| "loss": 0.4647, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.5225752508361204, | |
| "grad_norm": 2.48380970954895, | |
| "learning_rate": 9.755929367956354e-06, | |
| "loss": 0.48, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.5434782608695652, | |
| "grad_norm": 2.637362241744995, | |
| "learning_rate": 9.72257021361171e-06, | |
| "loss": 0.4795, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.56438127090301, | |
| "grad_norm": 2.8286116123199463, | |
| "learning_rate": 9.687139583888971e-06, | |
| "loss": 0.4685, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.5852842809364549, | |
| "grad_norm": 2.284113645553589, | |
| "learning_rate": 9.649653019833327e-06, | |
| "loss": 0.4574, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.6061872909698997, | |
| "grad_norm": 2.1332271099090576, | |
| "learning_rate": 9.610126964290714e-06, | |
| "loss": 0.4485, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.6270903010033445, | |
| "grad_norm": 2.4212398529052734, | |
| "learning_rate": 9.568578754695443e-06, | |
| "loss": 0.4579, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6270903010033445, | |
| "eval_loss": 0.45767197012901306, | |
| "eval_runtime": 826.252, | |
| "eval_samples_per_second": 4.714, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 0.3879739474296348, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6479933110367893, | |
| "grad_norm": 2.293464422225952, | |
| "learning_rate": 9.525026615465423e-06, | |
| "loss": 0.4512, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.6688963210702341, | |
| "grad_norm": 2.1372318267822266, | |
| "learning_rate": 9.479489650008331e-06, | |
| "loss": 0.4321, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.689799331103679, | |
| "grad_norm": 2.1387200355529785, | |
| "learning_rate": 9.431987832342228e-06, | |
| "loss": 0.4433, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.7107023411371237, | |
| "grad_norm": 2.39378023147583, | |
| "learning_rate": 9.382541998334293e-06, | |
| "loss": 0.4423, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.7316053511705686, | |
| "grad_norm": 2.0761451721191406, | |
| "learning_rate": 9.331173836561522e-06, | |
| "loss": 0.4439, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.7525083612040134, | |
| "grad_norm": 1.8935145139694214, | |
| "learning_rate": 9.277905878797401e-06, | |
| "loss": 0.439, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.7734113712374582, | |
| "grad_norm": 2.234813690185547, | |
| "learning_rate": 9.222761490128726e-06, | |
| "loss": 0.4268, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.794314381270903, | |
| "grad_norm": 2.1282079219818115, | |
| "learning_rate": 9.165764858706912e-06, | |
| "loss": 0.4315, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.8152173913043478, | |
| "grad_norm": 2.3695404529571533, | |
| "learning_rate": 9.106940985138251e-06, | |
| "loss": 0.4243, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.8361204013377926, | |
| "grad_norm": 2.079967498779297, | |
| "learning_rate": 9.04631567151784e-06, | |
| "loss": 0.4335, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.8361204013377926, | |
| "eval_loss": 0.42458781599998474, | |
| "eval_runtime": 798.4523, | |
| "eval_samples_per_second": 4.878, | |
| "eval_steps_per_second": 0.153, | |
| "eval_wer": 0.3473133286810886, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.8570234113712375, | |
| "grad_norm": 2.4322237968444824, | |
| "learning_rate": 8.983915510111918e-06, | |
| "loss": 0.4199, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.8779264214046822, | |
| "grad_norm": 2.022068977355957, | |
| "learning_rate": 8.91976787169362e-06, | |
| "loss": 0.4188, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.8988294314381271, | |
| "grad_norm": 2.305279016494751, | |
| "learning_rate": 8.85390089353726e-06, | |
| "loss": 0.4178, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.919732441471572, | |
| "grad_norm": 2.107536792755127, | |
| "learning_rate": 8.78634346707638e-06, | |
| "loss": 0.428, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.9406354515050167, | |
| "grad_norm": 2.2298972606658936, | |
| "learning_rate": 8.717125225231018e-06, | |
| "loss": 0.4128, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": 2.2876055240631104, | |
| "learning_rate": 8.646276529409729e-06, | |
| "loss": 0.3969, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.9824414715719063, | |
| "grad_norm": 2.3144311904907227, | |
| "learning_rate": 8.57382845619208e-06, | |
| "loss": 0.4114, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 1.0033444816053512, | |
| "grad_norm": 2.173213481903076, | |
| "learning_rate": 8.499812783697406e-06, | |
| "loss": 0.3942, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.024247491638796, | |
| "grad_norm": 2.3361010551452637, | |
| "learning_rate": 8.424261977645909e-06, | |
| "loss": 0.3257, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.0451505016722409, | |
| "grad_norm": 1.9306939840316772, | |
| "learning_rate": 8.347209177118101e-06, | |
| "loss": 0.3237, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0451505016722409, | |
| "eval_loss": 0.4114519953727722, | |
| "eval_runtime": 783.1554, | |
| "eval_samples_per_second": 4.973, | |
| "eval_steps_per_second": 0.156, | |
| "eval_wer": 0.3308676436380554, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0660535117056855, | |
| "grad_norm": 2.0871758460998535, | |
| "learning_rate": 8.268688180018911e-06, | |
| "loss": 0.3065, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 1.0869565217391304, | |
| "grad_norm": 1.6896426677703857, | |
| "learning_rate": 8.188733428252811e-06, | |
| "loss": 0.3181, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.1078595317725752, | |
| "grad_norm": 2.0144145488739014, | |
| "learning_rate": 8.107379992616453e-06, | |
| "loss": 0.3209, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 1.12876254180602, | |
| "grad_norm": 1.8922556638717651, | |
| "learning_rate": 8.024663557415466e-06, | |
| "loss": 0.3179, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.149665551839465, | |
| "grad_norm": 2.8480031490325928, | |
| "learning_rate": 7.940620404812129e-06, | |
| "loss": 0.3162, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 1.1705685618729098, | |
| "grad_norm": 2.6659786701202393, | |
| "learning_rate": 7.85528739891083e-06, | |
| "loss": 0.3111, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.1914715719063544, | |
| "grad_norm": 1.8330721855163574, | |
| "learning_rate": 7.768701969588237e-06, | |
| "loss": 0.3051, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 1.2123745819397993, | |
| "grad_norm": 2.5169472694396973, | |
| "learning_rate": 7.680902096075327e-06, | |
| "loss": 0.3218, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.2332775919732442, | |
| "grad_norm": 1.8183653354644775, | |
| "learning_rate": 7.591926290298435e-06, | |
| "loss": 0.3243, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 1.254180602006689, | |
| "grad_norm": 1.7857028245925903, | |
| "learning_rate": 7.501813579986655e-06, | |
| "loss": 0.306, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.254180602006689, | |
| "eval_loss": 0.3986068367958069, | |
| "eval_runtime": 761.2274, | |
| "eval_samples_per_second": 5.117, | |
| "eval_steps_per_second": 0.16, | |
| "eval_wer": 0.3142126075831589, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.2750836120401337, | |
| "grad_norm": 2.2187106609344482, | |
| "learning_rate": 7.4106034915529786e-06, | |
| "loss": 0.3168, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 1.2959866220735785, | |
| "grad_norm": 1.7013944387435913, | |
| "learning_rate": 7.318336032756717e-06, | |
| "loss": 0.3073, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.3168896321070234, | |
| "grad_norm": 2.3944790363311768, | |
| "learning_rate": 7.225051675154768e-06, | |
| "loss": 0.315, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 1.3377926421404682, | |
| "grad_norm": 1.8916776180267334, | |
| "learning_rate": 7.13079133634945e-06, | |
| "loss": 0.3081, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.358695652173913, | |
| "grad_norm": 1.776918888092041, | |
| "learning_rate": 7.035596362040697e-06, | |
| "loss": 0.3006, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 1.379598662207358, | |
| "grad_norm": 1.7980273962020874, | |
| "learning_rate": 6.939508507890464e-06, | |
| "loss": 0.3165, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.4005016722408028, | |
| "grad_norm": 1.5471354722976685, | |
| "learning_rate": 6.842569921207314e-06, | |
| "loss": 0.3089, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 1.4214046822742474, | |
| "grad_norm": 1.9330084323883057, | |
| "learning_rate": 6.744823122459217e-06, | |
| "loss": 0.3091, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.4423076923076923, | |
| "grad_norm": 2.163132429122925, | |
| "learning_rate": 6.6463109866226675e-06, | |
| "loss": 0.3141, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 1.4632107023411371, | |
| "grad_norm": 2.1405696868896484, | |
| "learning_rate": 6.547076724376296e-06, | |
| "loss": 0.3247, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.4632107023411371, | |
| "eval_loss": 0.39163750410079956, | |
| "eval_runtime": 753.6708, | |
| "eval_samples_per_second": 5.168, | |
| "eval_steps_per_second": 0.162, | |
| "eval_wer": 0.3070714119562689, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.484113712374582, | |
| "grad_norm": 2.1151435375213623, | |
| "learning_rate": 6.447163863147251e-06, | |
| "loss": 0.3089, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 1.5050167224080266, | |
| "grad_norm": 1.852047085762024, | |
| "learning_rate": 6.3466162280186164e-06, | |
| "loss": 0.3063, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.5259197324414715, | |
| "grad_norm": 3.6557390689849854, | |
| "learning_rate": 6.24547792250629e-06, | |
| "loss": 0.3033, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 1.5468227424749164, | |
| "grad_norm": 1.9076846837997437, | |
| "learning_rate": 6.143793309213724e-06, | |
| "loss": 0.2988, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 1.5677257525083612, | |
| "grad_norm": 1.8795961141586304, | |
| "learning_rate": 6.041606990373012e-06, | |
| "loss": 0.308, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 1.588628762541806, | |
| "grad_norm": 2.3081769943237305, | |
| "learning_rate": 5.938963788280889e-06, | |
| "loss": 0.3007, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.609531772575251, | |
| "grad_norm": 2.8769354820251465, | |
| "learning_rate": 5.835908725638186e-06, | |
| "loss": 0.2965, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 1.6304347826086958, | |
| "grad_norm": 2.2705342769622803, | |
| "learning_rate": 5.7324870058013736e-06, | |
| "loss": 0.3008, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 1.6513377926421404, | |
| "grad_norm": 1.9445313215255737, | |
| "learning_rate": 5.628743992954896e-06, | |
| "loss": 0.2978, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 1.6722408026755853, | |
| "grad_norm": 1.9228079319000244, | |
| "learning_rate": 5.524725192212924e-06, | |
| "loss": 0.3082, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.6722408026755853, | |
| "eval_loss": 0.38020968437194824, | |
| "eval_runtime": 779.2722, | |
| "eval_samples_per_second": 4.998, | |
| "eval_steps_per_second": 0.157, | |
| "eval_wer": 0.31256106071179346, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.6931438127090301, | |
| "grad_norm": 5.912314414978027, | |
| "learning_rate": 5.420476229659319e-06, | |
| "loss": 0.3098, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 1.7140468227424748, | |
| "grad_norm": 1.9327845573425293, | |
| "learning_rate": 5.316042832334516e-06, | |
| "loss": 0.2991, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 1.7349498327759196, | |
| "grad_norm": 1.669538140296936, | |
| "learning_rate": 5.211470808178137e-06, | |
| "loss": 0.2974, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 1.7558528428093645, | |
| "grad_norm": 2.050136089324951, | |
| "learning_rate": 5.1068060259361155e-06, | |
| "loss": 0.293, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.7767558528428093, | |
| "grad_norm": 1.8757737874984741, | |
| "learning_rate": 5.002094395041147e-06, | |
| "loss": 0.2986, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 1.7976588628762542, | |
| "grad_norm": 1.9177494049072266, | |
| "learning_rate": 4.897381845475294e-06, | |
| "loss": 0.2945, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 1.818561872909699, | |
| "grad_norm": 1.9221559762954712, | |
| "learning_rate": 4.792714307623574e-06, | |
| "loss": 0.2899, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 1.839464882943144, | |
| "grad_norm": 2.0715246200561523, | |
| "learning_rate": 4.688137692127389e-06, | |
| "loss": 0.2897, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.8603678929765888, | |
| "grad_norm": 1.8086769580841064, | |
| "learning_rate": 4.583697869746582e-06, | |
| "loss": 0.2916, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 1.8812709030100334, | |
| "grad_norm": 1.7725239992141724, | |
| "learning_rate": 4.4794406512390175e-06, | |
| "loss": 0.2985, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.8812709030100334, | |
| "eval_loss": 0.36925992369651794, | |
| "eval_runtime": 830.7968, | |
| "eval_samples_per_second": 4.688, | |
| "eval_steps_per_second": 0.147, | |
| "eval_wer": 0.3385903698534543, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.9021739130434783, | |
| "grad_norm": 2.940992832183838, | |
| "learning_rate": 4.375411767266468e-06, | |
| "loss": 0.2974, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 1.84628164768219, | |
| "learning_rate": 4.2716568483356295e-06, | |
| "loss": 0.3018, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.9439799331103678, | |
| "grad_norm": 3.159107208251953, | |
| "learning_rate": 4.168221404783076e-06, | |
| "loss": 0.2889, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 1.9648829431438126, | |
| "grad_norm": 2.257148504257202, | |
| "learning_rate": 4.0651508068129264e-06, | |
| "loss": 0.291, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 1.9857859531772575, | |
| "grad_norm": 1.9802433252334595, | |
| "learning_rate": 3.962490264595961e-06, | |
| "loss": 0.2912, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 2.0066889632107023, | |
| "grad_norm": 14.395709991455078, | |
| "learning_rate": 3.860284808438962e-06, | |
| "loss": 0.2704, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.027591973244147, | |
| "grad_norm": 2.082376480102539, | |
| "learning_rate": 3.7585792690329224e-06, | |
| "loss": 0.2113, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 2.048494983277592, | |
| "grad_norm": 2.4334142208099365, | |
| "learning_rate": 3.657418257788822e-06, | |
| "loss": 0.2041, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 2.069397993311037, | |
| "grad_norm": 1.8355016708374023, | |
| "learning_rate": 3.556846147269598e-06, | |
| "loss": 0.2133, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 2.0903010033444818, | |
| "grad_norm": 1.7609572410583496, | |
| "learning_rate": 3.4569070517268616e-06, | |
| "loss": 0.2134, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0903010033444818, | |
| "eval_loss": 0.37735578417778015, | |
| "eval_runtime": 769.814, | |
| "eval_samples_per_second": 5.06, | |
| "eval_steps_per_second": 0.158, | |
| "eval_wer": 0.2979995347755292, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.1112040133779266, | |
| "grad_norm": 5.835057735443115, | |
| "learning_rate": 3.3576448077509373e-06, | |
| "loss": 0.2048, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 2.132107023411371, | |
| "grad_norm": 1.5528628826141357, | |
| "learning_rate": 3.25910295504268e-06, | |
| "loss": 0.2035, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 2.153010033444816, | |
| "grad_norm": 1.694047451019287, | |
| "learning_rate": 3.1613247173155247e-06, | |
| "loss": 0.2017, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 2.1739130434782608, | |
| "grad_norm": 4.0651984214782715, | |
| "learning_rate": 3.0643529833361425e-06, | |
| "loss": 0.2026, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 2.1948160535117056, | |
| "grad_norm": 1.7460774183273315, | |
| "learning_rate": 2.968230288112002e-06, | |
| "loss": 0.2025, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 2.2157190635451505, | |
| "grad_norm": 1.6023117303848267, | |
| "learning_rate": 2.872998794234111e-06, | |
| "loss": 0.2106, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 2.2366220735785953, | |
| "grad_norm": 1.591112494468689, | |
| "learning_rate": 2.7787002733831166e-06, | |
| "loss": 0.203, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 2.25752508361204, | |
| "grad_norm": 1.9378623962402344, | |
| "learning_rate": 2.6853760880068587e-06, | |
| "loss": 0.21, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 2.278428093645485, | |
| "grad_norm": 2.3009631633758545, | |
| "learning_rate": 2.593067173177436e-06, | |
| "loss": 0.2088, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 2.29933110367893, | |
| "grad_norm": 1.7757341861724854, | |
| "learning_rate": 2.501814018635732e-06, | |
| "loss": 0.2057, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.29933110367893, | |
| "eval_loss": 0.3728334307670593, | |
| "eval_runtime": 771.123, | |
| "eval_samples_per_second": 5.051, | |
| "eval_steps_per_second": 0.158, | |
| "eval_wer": 0.2964642940218655, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.3202341137123748, | |
| "grad_norm": 6.916594982147217, | |
| "learning_rate": 2.4116566510312734e-06, | |
| "loss": 0.2078, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 2.3411371237458196, | |
| "grad_norm": 2.954857110977173, | |
| "learning_rate": 2.3226346163652213e-06, | |
| "loss": 0.2061, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 2.362040133779264, | |
| "grad_norm": 2.285639524459839, | |
| "learning_rate": 2.234786962644181e-06, | |
| "loss": 0.2066, | |
| "step": 5650 | |
| }, | |
| { | |
| "epoch": 2.382943143812709, | |
| "grad_norm": 2.133755683898926, | |
| "learning_rate": 2.1481522227524725e-06, | |
| "loss": 0.2034, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 2.4038461538461537, | |
| "grad_norm": 1.8175289630889893, | |
| "learning_rate": 2.0627683975503165e-06, | |
| "loss": 0.2056, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 2.4247491638795986, | |
| "grad_norm": 2.0304114818573, | |
| "learning_rate": 1.978672939205404e-06, | |
| "loss": 0.1999, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 2.4456521739130435, | |
| "grad_norm": 1.922811508178711, | |
| "learning_rate": 1.8959027347651527e-06, | |
| "loss": 0.1985, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 2.4665551839464883, | |
| "grad_norm": 1.7293944358825684, | |
| "learning_rate": 1.814494089976805e-06, | |
| "loss": 0.2045, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 2.487458193979933, | |
| "grad_norm": 1.9660613536834717, | |
| "learning_rate": 1.7344827133625487e-06, | |
| "loss": 0.1967, | |
| "step": 5950 | |
| }, | |
| { | |
| "epoch": 2.508361204013378, | |
| "grad_norm": 1.8475674390792847, | |
| "learning_rate": 1.6559037005565665e-06, | |
| "loss": 0.208, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.508361204013378, | |
| "eval_loss": 0.37195536494255066, | |
| "eval_runtime": 759.7026, | |
| "eval_samples_per_second": 5.127, | |
| "eval_steps_per_second": 0.161, | |
| "eval_wer": 0.28855547801814374, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.529264214046823, | |
| "grad_norm": 1.9271501302719116, | |
| "learning_rate": 1.578791518910937e-06, | |
| "loss": 0.2095, | |
| "step": 6050 | |
| }, | |
| { | |
| "epoch": 2.5501672240802673, | |
| "grad_norm": 1.8744512796401978, | |
| "learning_rate": 1.5031799923771102e-06, | |
| "loss": 0.2045, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 2.571070234113712, | |
| "grad_norm": 1.887246012687683, | |
| "learning_rate": 1.4291022866696086e-06, | |
| "loss": 0.211, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 2.591973244147157, | |
| "grad_norm": 1.6180899143218994, | |
| "learning_rate": 1.356590894718447e-06, | |
| "loss": 0.1994, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 2.612876254180602, | |
| "grad_norm": 1.6288968324661255, | |
| "learning_rate": 1.2856776224166589e-06, | |
| "loss": 0.1948, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 2.6337792642140467, | |
| "grad_norm": 3.4187662601470947, | |
| "learning_rate": 1.2163935746691807e-06, | |
| "loss": 0.1932, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 2.6546822742474916, | |
| "grad_norm": 1.5604482889175415, | |
| "learning_rate": 1.1487691417492147e-06, | |
| "loss": 0.2028, | |
| "step": 6350 | |
| }, | |
| { | |
| "epoch": 2.6755852842809364, | |
| "grad_norm": 1.7518872022628784, | |
| "learning_rate": 1.0828339859680487e-06, | |
| "loss": 0.1984, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 2.6964882943143813, | |
| "grad_norm": 1.7024357318878174, | |
| "learning_rate": 1.0186170286641816e-06, | |
| "loss": 0.2007, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 2.717391304347826, | |
| "grad_norm": 1.413225531578064, | |
| "learning_rate": 9.561464375174827e-07, | |
| "loss": 0.1998, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.717391304347826, | |
| "eval_loss": 0.3693811297416687, | |
| "eval_runtime": 748.5143, | |
| "eval_samples_per_second": 5.204, | |
| "eval_steps_per_second": 0.163, | |
| "eval_wer": 0.2801814375436148, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.738294314381271, | |
| "grad_norm": 1.974918007850647, | |
| "learning_rate": 8.954496141938973e-07, | |
| "loss": 0.2044, | |
| "step": 6550 | |
| }, | |
| { | |
| "epoch": 2.759197324414716, | |
| "grad_norm": 1.8519419431686401, | |
| "learning_rate": 8.365531823261841e-07, | |
| "loss": 0.1974, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 2.7801003344481607, | |
| "grad_norm": 1.467334508895874, | |
| "learning_rate": 7.794829758359085e-07, | |
| "loss": 0.1947, | |
| "step": 6650 | |
| }, | |
| { | |
| "epoch": 2.8010033444816056, | |
| "grad_norm": 1.6851749420166016, | |
| "learning_rate": 7.242640276018226e-07, | |
| "loss": 0.1903, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 2.82190635451505, | |
| "grad_norm": 1.7586499452590942, | |
| "learning_rate": 6.709205584796241e-07, | |
| "loss": 0.2005, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 2.842809364548495, | |
| "grad_norm": 1.6368706226348877, | |
| "learning_rate": 6.194759666778927e-07, | |
| "loss": 0.2022, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 2.8637123745819397, | |
| "grad_norm": 1.6208401918411255, | |
| "learning_rate": 5.699528174948477e-07, | |
| "loss": 0.1998, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 2.8846153846153846, | |
| "grad_norm": 1.6510974168777466, | |
| "learning_rate": 5.22372833420478e-07, | |
| "loss": 0.1993, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 2.9055183946488294, | |
| "grad_norm": 1.4576036930084229, | |
| "learning_rate": 4.7675688460833145e-07, | |
| "loss": 0.1899, | |
| "step": 6950 | |
| }, | |
| { | |
| "epoch": 2.9264214046822743, | |
| "grad_norm": 1.6056406497955322, | |
| "learning_rate": 4.3312497972117895e-07, | |
| "loss": 0.201, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.9264214046822743, | |
| "eval_loss": 0.36715638637542725, | |
| "eval_runtime": 761.783, | |
| "eval_samples_per_second": 5.113, | |
| "eval_steps_per_second": 0.16, | |
| "eval_wer": 0.28343800883926495, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.947324414715719, | |
| "grad_norm": 1.867806315422058, | |
| "learning_rate": 3.9149625715455107e-07, | |
| "loss": 0.1963, | |
| "step": 7050 | |
| }, | |
| { | |
| "epoch": 2.968227424749164, | |
| "grad_norm": 2.8958945274353027, | |
| "learning_rate": 3.5188897664200804e-07, | |
| "loss": 0.1962, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 2.9891304347826084, | |
| "grad_norm": 1.6264315843582153, | |
| "learning_rate": 3.14320511245817e-07, | |
| "loss": 0.2003, | |
| "step": 7150 | |
| }, | |
| { | |
| "epoch": 3.0100334448160537, | |
| "grad_norm": 1.27947998046875, | |
| "learning_rate": 2.788073397365465e-07, | |
| "loss": 0.1805, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 3.030936454849498, | |
| "grad_norm": 1.8425559997558594, | |
| "learning_rate": 2.453650393649448e-07, | |
| "loss": 0.1597, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 3.051839464882943, | |
| "grad_norm": 1.5015190839767456, | |
| "learning_rate": 2.1400827902923304e-07, | |
| "loss": 0.16, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 3.072742474916388, | |
| "grad_norm": 1.3788676261901855, | |
| "learning_rate": 1.8475081284084428e-07, | |
| "loss": 0.1648, | |
| "step": 7350 | |
| }, | |
| { | |
| "epoch": 3.0936454849498327, | |
| "grad_norm": 1.9593479633331299, | |
| "learning_rate": 1.5760547409141626e-07, | |
| "loss": 0.1562, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 3.1145484949832776, | |
| "grad_norm": 1.521888256072998, | |
| "learning_rate": 1.3258416962368849e-07, | |
| "loss": 0.1616, | |
| "step": 7450 | |
| }, | |
| { | |
| "epoch": 3.1354515050167224, | |
| "grad_norm": 3.2412497997283936, | |
| "learning_rate": 1.0969787460876013e-07, | |
| "loss": 0.1601, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.1354515050167224, | |
| "eval_loss": 0.3784402906894684, | |
| "eval_runtime": 762.4332, | |
| "eval_samples_per_second": 5.109, | |
| "eval_steps_per_second": 0.16, | |
| "eval_wer": 0.2860200046522447, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.1563545150501673, | |
| "grad_norm": 1.6288435459136963, | |
| "learning_rate": 8.895662773202529e-08, | |
| "loss": 0.1571, | |
| "step": 7550 | |
| }, | |
| { | |
| "epoch": 3.177257525083612, | |
| "grad_norm": 1.7533422708511353, | |
| "learning_rate": 7.036952678986852e-08, | |
| "loss": 0.1572, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 3.198160535117057, | |
| "grad_norm": 2.158843994140625, | |
| "learning_rate": 5.394472469907208e-08, | |
| "loss": 0.1566, | |
| "step": 7650 | |
| }, | |
| { | |
| "epoch": 3.219063545150502, | |
| "grad_norm": 1.5002875328063965, | |
| "learning_rate": 3.9689425920678146e-08, | |
| "loss": 0.1618, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 3.2399665551839467, | |
| "grad_norm": 1.6574152708053589, | |
| "learning_rate": 2.760988329987224e-08, | |
| "loss": 0.1581, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 3.260869565217391, | |
| "grad_norm": 1.6822019815444946, | |
| "learning_rate": 1.7711395323281588e-08, | |
| "loss": 0.1601, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 3.281772575250836, | |
| "grad_norm": 1.6592892408370972, | |
| "learning_rate": 9.99830379487976e-09, | |
| "loss": 0.1534, | |
| "step": 7850 | |
| }, | |
| { | |
| "epoch": 3.302675585284281, | |
| "grad_norm": 1.4634497165679932, | |
| "learning_rate": 4.473991931531752e-09, | |
| "loss": 0.1623, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 3.3235785953177257, | |
| "grad_norm": 1.8965901136398315, | |
| "learning_rate": 1.1408828790010484e-09, | |
| "loss": 0.161, | |
| "step": 7950 | |
| }, | |
| { | |
| "epoch": 3.3444816053511706, | |
| "grad_norm": 1.7498852014541626, | |
| "learning_rate": 4.386490781715935e-13, | |
| "loss": 0.1591, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.3444816053511706, | |
| "eval_loss": 0.37963372468948364, | |
| "eval_runtime": 754.8076, | |
| "eval_samples_per_second": 5.16, | |
| "eval_steps_per_second": 0.162, | |
| "eval_wer": 0.28353105373342635, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.3444816053511706, | |
| "step": 8000, | |
| "total_flos": 1.0870635728528317e+21, | |
| "train_loss": 0.3253028565645218, | |
| "train_runtime": 55615.1831, | |
| "train_samples_per_second": 9.206, | |
| "train_steps_per_second": 0.144 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 8000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 2 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0870635728528317e+21, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |