| { | |
| "best_metric": 0.09467801134467801, | |
| "best_model_checkpoint": "fine-w2v2base-bs16-ep100-lr2e-05-linguistic-rmsnorm-focal_ctc_a0.75_g1.0-0.05_10_0.004_40/checkpoint-3150", | |
| "epoch": 100.0, | |
| "eval_steps": 50, | |
| "global_step": 5300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 7.547169811320755e-07, | |
| "loss": 1628.3266, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_loss": 795.486328125, | |
| "eval_runtime": 3.3117, | |
| "eval_samples_per_second": 216.507, | |
| "eval_steps_per_second": 6.945, | |
| "eval_wer": 15.705538872205539, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 2.5660377358490568e-06, | |
| "loss": 1167.4639, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 276.5289001464844, | |
| "eval_runtime": 2.3138, | |
| "eval_samples_per_second": 309.881, | |
| "eval_steps_per_second": 9.94, | |
| "eval_wer": 0.995412078745412, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 4.452830188679246e-06, | |
| "loss": 193.5269, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "eval_loss": 67.26387786865234, | |
| "eval_runtime": 2.2447, | |
| "eval_samples_per_second": 319.421, | |
| "eval_steps_per_second": 10.246, | |
| "eval_wer": 1.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 6.339622641509434e-06, | |
| "loss": 86.2802, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "eval_loss": 63.62594985961914, | |
| "eval_runtime": 2.2662, | |
| "eval_samples_per_second": 316.387, | |
| "eval_steps_per_second": 10.149, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 8.226415094339623e-06, | |
| "loss": 82.8944, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "eval_loss": 61.89385986328125, | |
| "eval_runtime": 2.2689, | |
| "eval_samples_per_second": 316.007, | |
| "eval_steps_per_second": 10.137, | |
| "eval_wer": 1.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 1.0113207547169812e-05, | |
| "loss": 79.9575, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "eval_loss": 59.79137420654297, | |
| "eval_runtime": 2.2608, | |
| "eval_samples_per_second": 317.142, | |
| "eval_steps_per_second": 10.173, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "learning_rate": 1.2e-05, | |
| "loss": 77.3429, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "eval_loss": 58.15074157714844, | |
| "eval_runtime": 2.2539, | |
| "eval_samples_per_second": 318.116, | |
| "eval_steps_per_second": 10.205, | |
| "eval_wer": 1.0, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 7.55, | |
| "learning_rate": 1.3886792452830189e-05, | |
| "loss": 74.0584, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.55, | |
| "eval_loss": 57.29766082763672, | |
| "eval_runtime": 2.2566, | |
| "eval_samples_per_second": 317.741, | |
| "eval_steps_per_second": 10.193, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 1.577358490566038e-05, | |
| "loss": 72.6694, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "eval_loss": 56.71151351928711, | |
| "eval_runtime": 2.3841, | |
| "eval_samples_per_second": 300.746, | |
| "eval_steps_per_second": 9.647, | |
| "eval_wer": 1.0, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "learning_rate": 1.766037735849057e-05, | |
| "loss": 73.417, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "eval_loss": 56.60741424560547, | |
| "eval_runtime": 2.3092, | |
| "eval_samples_per_second": 310.501, | |
| "eval_steps_per_second": 9.96, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 1.9547169811320757e-05, | |
| "loss": 72.4291, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "eval_loss": 56.47547912597656, | |
| "eval_runtime": 2.313, | |
| "eval_samples_per_second": 309.989, | |
| "eval_steps_per_second": 9.944, | |
| "eval_wer": 1.0, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 11.32, | |
| "learning_rate": 1.9996868319012422e-05, | |
| "loss": 72.7847, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 11.32, | |
| "eval_loss": 56.86225128173828, | |
| "eval_runtime": 2.2819, | |
| "eval_samples_per_second": 314.217, | |
| "eval_steps_per_second": 10.079, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "learning_rate": 1.9983208992285993e-05, | |
| "loss": 69.3297, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "eval_loss": 49.65397644042969, | |
| "eval_runtime": 2.2885, | |
| "eval_samples_per_second": 313.303, | |
| "eval_steps_per_second": 10.05, | |
| "eval_wer": 0.9637137137137137, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 13.21, | |
| "learning_rate": 1.9958724515842856e-05, | |
| "loss": 54.2644, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 13.21, | |
| "eval_loss": 29.155927658081055, | |
| "eval_runtime": 2.3174, | |
| "eval_samples_per_second": 309.402, | |
| "eval_steps_per_second": 9.925, | |
| "eval_wer": 0.5630630630630631, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "learning_rate": 1.9923441439074434e-05, | |
| "loss": 31.2303, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "eval_loss": 13.895669937133789, | |
| "eval_runtime": 2.2929, | |
| "eval_samples_per_second": 312.699, | |
| "eval_steps_per_second": 10.031, | |
| "eval_wer": 0.24140807474140807, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 15.09, | |
| "learning_rate": 1.9877398020679796e-05, | |
| "loss": 19.4522, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 15.09, | |
| "eval_loss": 9.745962142944336, | |
| "eval_runtime": 2.3244, | |
| "eval_samples_per_second": 308.464, | |
| "eval_steps_per_second": 9.895, | |
| "eval_wer": 0.19494494494494494, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "learning_rate": 1.9820644187180354e-05, | |
| "loss": 15.0046, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "eval_loss": 7.673460483551025, | |
| "eval_runtime": 2.357, | |
| "eval_samples_per_second": 304.198, | |
| "eval_steps_per_second": 9.758, | |
| "eval_wer": 0.16216216216216217, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "learning_rate": 1.975324147878278e-05, | |
| "loss": 12.3783, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "eval_loss": 6.555938720703125, | |
| "eval_runtime": 2.2666, | |
| "eval_samples_per_second": 316.327, | |
| "eval_steps_per_second": 10.147, | |
| "eval_wer": 0.1519853186519853, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 17.92, | |
| "learning_rate": 1.9675262982648757e-05, | |
| "loss": 10.7256, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 17.92, | |
| "eval_loss": 5.785167217254639, | |
| "eval_runtime": 2.3021, | |
| "eval_samples_per_second": 311.456, | |
| "eval_steps_per_second": 9.991, | |
| "eval_wer": 0.142308975642309, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 18.87, | |
| "learning_rate": 1.958679325364396e-05, | |
| "loss": 9.8218, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 18.87, | |
| "eval_loss": 5.4473090171813965, | |
| "eval_runtime": 2.3201, | |
| "eval_samples_per_second": 309.032, | |
| "eval_steps_per_second": 9.913, | |
| "eval_wer": 0.1394728061394728, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 19.81, | |
| "learning_rate": 1.9487928222652195e-05, | |
| "loss": 9.0115, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 19.81, | |
| "eval_loss": 5.125023365020752, | |
| "eval_runtime": 2.3061, | |
| "eval_samples_per_second": 310.909, | |
| "eval_steps_per_second": 9.973, | |
| "eval_wer": 0.13563563563563563, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 20.75, | |
| "learning_rate": 1.9378775092554124e-05, | |
| "loss": 8.1076, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 20.75, | |
| "eval_loss": 4.797976016998291, | |
| "eval_runtime": 2.2582, | |
| "eval_samples_per_second": 317.516, | |
| "eval_steps_per_second": 10.185, | |
| "eval_wer": 0.12328995662328995, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 21.7, | |
| "learning_rate": 1.925945222198336e-05, | |
| "loss": 7.9779, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 21.7, | |
| "eval_loss": 4.615002632141113, | |
| "eval_runtime": 2.3017, | |
| "eval_samples_per_second": 311.506, | |
| "eval_steps_per_second": 9.993, | |
| "eval_wer": 0.12112112112112113, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 22.64, | |
| "learning_rate": 1.9130088996985967e-05, | |
| "loss": 7.6027, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 22.64, | |
| "eval_loss": 4.650717258453369, | |
| "eval_runtime": 2.245, | |
| "eval_samples_per_second": 319.376, | |
| "eval_steps_per_second": 10.245, | |
| "eval_wer": 0.12512512512512514, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 23.58, | |
| "learning_rate": 1.8990825690722557e-05, | |
| "loss": 7.4535, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 23.58, | |
| "eval_loss": 4.481402397155762, | |
| "eval_runtime": 2.3288, | |
| "eval_samples_per_second": 307.882, | |
| "eval_steps_per_second": 9.876, | |
| "eval_wer": 0.12095428762095428, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 24.53, | |
| "learning_rate": 1.8841813311365105e-05, | |
| "loss": 6.946, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 24.53, | |
| "eval_loss": 4.436919212341309, | |
| "eval_runtime": 2.254, | |
| "eval_samples_per_second": 318.102, | |
| "eval_steps_per_second": 10.204, | |
| "eval_wer": 0.11494828161494829, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 25.47, | |
| "learning_rate": 1.868321343835339e-05, | |
| "loss": 7.0627, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 25.47, | |
| "eval_loss": 4.11531400680542, | |
| "eval_runtime": 2.2753, | |
| "eval_samples_per_second": 315.126, | |
| "eval_steps_per_second": 10.109, | |
| "eval_wer": 0.11394728061394728, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 26.42, | |
| "learning_rate": 1.8515198047188652e-05, | |
| "loss": 6.2482, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 26.42, | |
| "eval_loss": 4.004451274871826, | |
| "eval_runtime": 2.232, | |
| "eval_samples_per_second": 321.232, | |
| "eval_steps_per_second": 10.305, | |
| "eval_wer": 0.11011011011011011, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 27.36, | |
| "learning_rate": 1.833794932295441e-05, | |
| "loss": 6.2238, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 27.36, | |
| "eval_loss": 4.035464286804199, | |
| "eval_runtime": 2.4345, | |
| "eval_samples_per_second": 294.516, | |
| "eval_steps_per_second": 9.448, | |
| "eval_wer": 0.11578244911578245, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 28.3, | |
| "learning_rate": 1.8151659462766685e-05, | |
| "loss": 5.8919, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 28.3, | |
| "eval_loss": 3.9624619483947754, | |
| "eval_runtime": 2.2497, | |
| "eval_samples_per_second": 318.712, | |
| "eval_steps_per_second": 10.224, | |
| "eval_wer": 0.11544878211544878, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 29.25, | |
| "learning_rate": 1.7956530467367805e-05, | |
| "loss": 5.7955, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 29.25, | |
| "eval_loss": 3.7957444190979004, | |
| "eval_runtime": 2.3687, | |
| "eval_samples_per_second": 302.701, | |
| "eval_steps_per_second": 9.71, | |
| "eval_wer": 0.11269602936269603, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 30.19, | |
| "learning_rate": 1.7752773922089784e-05, | |
| "loss": 5.4849, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 30.19, | |
| "eval_loss": 3.7985599040985107, | |
| "eval_runtime": 2.3776, | |
| "eval_samples_per_second": 301.568, | |
| "eval_steps_per_second": 9.674, | |
| "eval_wer": 0.10577243910577244, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 31.13, | |
| "learning_rate": 1.7540610767424813e-05, | |
| "loss": 5.1108, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 31.13, | |
| "eval_loss": 3.818775177001953, | |
| "eval_runtime": 2.2702, | |
| "eval_samples_per_second": 315.836, | |
| "eval_steps_per_second": 10.131, | |
| "eval_wer": 0.10702369035702369, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 32.08, | |
| "learning_rate": 1.7320271059451597e-05, | |
| "loss": 5.3354, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 32.08, | |
| "eval_loss": 3.69089674949646, | |
| "eval_runtime": 2.3064, | |
| "eval_samples_per_second": 310.88, | |
| "eval_steps_per_second": 9.972, | |
| "eval_wer": 0.10235235235235235, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 33.02, | |
| "learning_rate": 1.7091993720377336e-05, | |
| "loss": 5.1149, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 33.02, | |
| "eval_loss": 3.62273907661438, | |
| "eval_runtime": 2.3105, | |
| "eval_samples_per_second": 310.319, | |
| "eval_steps_per_second": 9.954, | |
| "eval_wer": 0.10226893560226893, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 33.96, | |
| "learning_rate": 1.685602627946584e-05, | |
| "loss": 4.976, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 33.96, | |
| "eval_loss": 3.6176178455352783, | |
| "eval_runtime": 2.4073, | |
| "eval_samples_per_second": 297.846, | |
| "eval_steps_per_second": 9.554, | |
| "eval_wer": 0.1016016016016016, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 34.91, | |
| "learning_rate": 1.661262460463274e-05, | |
| "loss": 4.5904, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 34.91, | |
| "eval_loss": 3.5958945751190186, | |
| "eval_runtime": 2.3015, | |
| "eval_samples_per_second": 311.541, | |
| "eval_steps_per_second": 9.994, | |
| "eval_wer": 0.10785785785785786, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 35.85, | |
| "learning_rate": 1.6362052624998767e-05, | |
| "loss": 4.6613, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 35.85, | |
| "eval_loss": 3.4999887943267822, | |
| "eval_runtime": 2.265, | |
| "eval_samples_per_second": 316.562, | |
| "eval_steps_per_second": 10.155, | |
| "eval_wer": 0.10685685685685686, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 36.79, | |
| "learning_rate": 1.6104582044701983e-05, | |
| "loss": 4.7697, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 36.79, | |
| "eval_loss": 3.5211145877838135, | |
| "eval_runtime": 2.326, | |
| "eval_samples_per_second": 308.248, | |
| "eval_steps_per_second": 9.888, | |
| "eval_wer": 0.10143476810143477, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 37.74, | |
| "learning_rate": 1.584049204827929e-05, | |
| "loss": 4.4224, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 37.74, | |
| "eval_loss": 3.4720044136047363, | |
| "eval_runtime": 2.25, | |
| "eval_samples_per_second": 318.672, | |
| "eval_steps_per_second": 10.222, | |
| "eval_wer": 0.1001001001001001, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 38.68, | |
| "learning_rate": 1.5570068997936686e-05, | |
| "loss": 4.5255, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 38.68, | |
| "eval_loss": 3.417844533920288, | |
| "eval_runtime": 2.2907, | |
| "eval_samples_per_second": 313.008, | |
| "eval_steps_per_second": 10.041, | |
| "eval_wer": 0.09826493159826494, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 39.62, | |
| "learning_rate": 1.5293606123036508e-05, | |
| "loss": 4.2808, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 39.62, | |
| "eval_loss": 3.4800703525543213, | |
| "eval_runtime": 2.3337, | |
| "eval_samples_per_second": 307.243, | |
| "eval_steps_per_second": 9.856, | |
| "eval_wer": 0.10435435435435435, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 40.57, | |
| "learning_rate": 1.5011403202138346e-05, | |
| "loss": 4.2407, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 40.57, | |
| "eval_loss": 3.4080042839050293, | |
| "eval_runtime": 2.2974, | |
| "eval_samples_per_second": 312.092, | |
| "eval_steps_per_second": 10.011, | |
| "eval_wer": 0.10001668335001668, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 41.51, | |
| "learning_rate": 1.4723766237938495e-05, | |
| "loss": 3.9611, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 41.51, | |
| "eval_loss": 3.451436758041382, | |
| "eval_runtime": 2.3789, | |
| "eval_samples_per_second": 301.4, | |
| "eval_steps_per_second": 9.668, | |
| "eval_wer": 0.10485485485485485, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 42.45, | |
| "learning_rate": 1.4431007125460274e-05, | |
| "loss": 4.014, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 42.45, | |
| "eval_loss": 3.3982768058776855, | |
| "eval_runtime": 2.2505, | |
| "eval_samples_per_second": 318.595, | |
| "eval_steps_per_second": 10.22, | |
| "eval_wer": 0.10885885885885886, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 43.4, | |
| "learning_rate": 1.4133443313855155e-05, | |
| "loss": 3.8487, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 43.4, | |
| "eval_loss": 3.4163670539855957, | |
| "eval_runtime": 2.4151, | |
| "eval_samples_per_second": 296.887, | |
| "eval_steps_per_second": 9.524, | |
| "eval_wer": 0.10418752085418752, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 44.34, | |
| "learning_rate": 1.3831397462181298e-05, | |
| "loss": 3.8132, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 44.34, | |
| "eval_loss": 3.356208086013794, | |
| "eval_runtime": 2.3331, | |
| "eval_samples_per_second": 307.322, | |
| "eval_steps_per_second": 9.858, | |
| "eval_wer": 0.09584584584584585, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 45.28, | |
| "learning_rate": 1.3525197089532833e-05, | |
| "loss": 3.6973, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 45.28, | |
| "eval_loss": 3.283907890319824, | |
| "eval_runtime": 2.2596, | |
| "eval_samples_per_second": 317.308, | |
| "eval_steps_per_second": 10.179, | |
| "eval_wer": 0.09776443109776443, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 46.23, | |
| "learning_rate": 1.3215174219899224e-05, | |
| "loss": 3.606, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 46.23, | |
| "eval_loss": 3.3125367164611816, | |
| "eval_runtime": 2.2757, | |
| "eval_samples_per_second": 315.07, | |
| "eval_steps_per_second": 10.107, | |
| "eval_wer": 0.10093426760093427, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 47.17, | |
| "learning_rate": 1.2901665022139796e-05, | |
| "loss": 3.5412, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 47.17, | |
| "eval_loss": 3.2580113410949707, | |
| "eval_runtime": 2.2944, | |
| "eval_samples_per_second": 312.502, | |
| "eval_steps_per_second": 10.024, | |
| "eval_wer": 0.09768101434768102, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 48.11, | |
| "learning_rate": 1.2591371175914464e-05, | |
| "loss": 3.3971, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 48.11, | |
| "eval_loss": 3.306504487991333, | |
| "eval_runtime": 2.3284, | |
| "eval_samples_per_second": 307.938, | |
| "eval_steps_per_second": 9.878, | |
| "eval_wer": 0.09843176509843177, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 49.06, | |
| "learning_rate": 1.2271965256408201e-05, | |
| "loss": 3.4795, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 49.06, | |
| "eval_loss": 3.3312034606933594, | |
| "eval_runtime": 2.3017, | |
| "eval_samples_per_second": 311.51, | |
| "eval_steps_per_second": 9.993, | |
| "eval_wer": 0.10368702035368701, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 1.1950095763909446e-05, | |
| "loss": 3.302, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 3.3014612197875977, | |
| "eval_runtime": 2.4125, | |
| "eval_samples_per_second": 297.198, | |
| "eval_steps_per_second": 9.534, | |
| "eval_wer": 0.0985985985985986, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 50.94, | |
| "learning_rate": 1.1626111712993397e-05, | |
| "loss": 3.2486, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 50.94, | |
| "eval_loss": 3.2506401538848877, | |
| "eval_runtime": 2.2778, | |
| "eval_samples_per_second": 314.775, | |
| "eval_steps_per_second": 10.097, | |
| "eval_wer": 0.09768101434768102, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 51.89, | |
| "learning_rate": 1.1300364411126395e-05, | |
| "loss": 3.3977, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 51.89, | |
| "eval_loss": 3.240551471710205, | |
| "eval_runtime": 2.4135, | |
| "eval_samples_per_second": 297.074, | |
| "eval_steps_per_second": 9.53, | |
| "eval_wer": 0.09517851184517852, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 52.83, | |
| "learning_rate": 1.0973207077730688e-05, | |
| "loss": 3.0229, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 52.83, | |
| "eval_loss": 3.287993907928467, | |
| "eval_runtime": 2.3151, | |
| "eval_samples_per_second": 309.7, | |
| "eval_steps_per_second": 9.935, | |
| "eval_wer": 0.09893226559893227, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 53.77, | |
| "learning_rate": 1.0644994461176039e-05, | |
| "loss": 3.2615, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 53.77, | |
| "eval_loss": 3.3112218379974365, | |
| "eval_runtime": 2.2897, | |
| "eval_samples_per_second": 313.137, | |
| "eval_steps_per_second": 10.045, | |
| "eval_wer": 0.09984984984984985, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 54.72, | |
| "learning_rate": 1.0316082454113379e-05, | |
| "loss": 3.2023, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 54.72, | |
| "eval_loss": 3.289529323577881, | |
| "eval_runtime": 2.3096, | |
| "eval_samples_per_second": 310.45, | |
| "eval_steps_per_second": 9.959, | |
| "eval_wer": 0.10368702035368701, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 55.66, | |
| "learning_rate": 9.98682770756773e-06, | |
| "loss": 3.0037, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 55.66, | |
| "eval_loss": 3.33939266204834, | |
| "eval_runtime": 2.3223, | |
| "eval_samples_per_second": 308.74, | |
| "eval_steps_per_second": 9.904, | |
| "eval_wer": 0.10176843510176843, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 56.6, | |
| "learning_rate": 9.657587244208758e-06, | |
| "loss": 2.9249, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 56.6, | |
| "eval_loss": 3.2351438999176025, | |
| "eval_runtime": 2.3103, | |
| "eval_samples_per_second": 310.353, | |
| "eval_steps_per_second": 9.956, | |
| "eval_wer": 0.09743076409743076, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 57.55, | |
| "learning_rate": 9.328718071218366e-06, | |
| "loss": 3.112, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 57.55, | |
| "eval_loss": 3.2868096828460693, | |
| "eval_runtime": 2.2583, | |
| "eval_samples_per_second": 317.501, | |
| "eval_steps_per_second": 10.185, | |
| "eval_wer": 0.10185185185185185, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 58.49, | |
| "learning_rate": 9.000576793175061e-06, | |
| "loss": 3.0261, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 58.49, | |
| "eval_loss": 3.324059009552002, | |
| "eval_runtime": 2.3538, | |
| "eval_samples_per_second": 304.612, | |
| "eval_steps_per_second": 9.771, | |
| "eval_wer": 0.10385385385385386, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 59.43, | |
| "learning_rate": 8.673519225374882e-06, | |
| "loss": 2.8959, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 59.43, | |
| "eval_loss": 3.225084066390991, | |
| "eval_runtime": 2.3969, | |
| "eval_samples_per_second": 299.14, | |
| "eval_steps_per_second": 9.596, | |
| "eval_wer": 0.09467801134467801, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 60.38, | |
| "learning_rate": 8.347900008008194e-06, | |
| "loss": 2.946, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 60.38, | |
| "eval_loss": 3.28798508644104, | |
| "eval_runtime": 2.4104, | |
| "eval_samples_per_second": 297.466, | |
| "eval_steps_per_second": 9.542, | |
| "eval_wer": 0.10118451785118451, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 61.32, | |
| "learning_rate": 8.024072221610653e-06, | |
| "loss": 2.6933, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 61.32, | |
| "eval_loss": 3.259455680847168, | |
| "eval_runtime": 2.3334, | |
| "eval_samples_per_second": 307.279, | |
| "eval_steps_per_second": 9.857, | |
| "eval_wer": 0.1031031031031031, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 62.26, | |
| "learning_rate": 7.702387004205407e-06, | |
| "loss": 2.8755, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 62.26, | |
| "eval_loss": 3.214036464691162, | |
| "eval_runtime": 2.2931, | |
| "eval_samples_per_second": 312.677, | |
| "eval_steps_per_second": 10.03, | |
| "eval_wer": 0.10477143810477144, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 63.21, | |
| "learning_rate": 7.383193170551595e-06, | |
| "loss": 2.606, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 63.21, | |
| "eval_loss": 3.2743003368377686, | |
| "eval_runtime": 2.33, | |
| "eval_samples_per_second": 307.725, | |
| "eval_steps_per_second": 9.871, | |
| "eval_wer": 0.1075241908575242, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 64.15, | |
| "learning_rate": 7.066836833912053e-06, | |
| "loss": 2.7607, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 64.15, | |
| "eval_loss": 3.2454757690429688, | |
| "eval_runtime": 2.3209, | |
| "eval_samples_per_second": 308.936, | |
| "eval_steps_per_second": 9.91, | |
| "eval_wer": 0.10527193860527194, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 65.09, | |
| "learning_rate": 6.7536610307503735e-06, | |
| "loss": 2.6394, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 65.09, | |
| "eval_loss": 3.2334530353546143, | |
| "eval_runtime": 2.3834, | |
| "eval_samples_per_second": 300.837, | |
| "eval_steps_per_second": 9.65, | |
| "eval_wer": 0.09943276609943276, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 66.04, | |
| "learning_rate": 6.444005348764207e-06, | |
| "loss": 2.6899, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 66.04, | |
| "eval_loss": 3.227813482284546, | |
| "eval_runtime": 2.2591, | |
| "eval_samples_per_second": 317.385, | |
| "eval_steps_per_second": 10.181, | |
| "eval_wer": 0.10043376710043377, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 66.98, | |
| "learning_rate": 6.138205558658212e-06, | |
| "loss": 2.719, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 66.98, | |
| "eval_loss": 3.201153516769409, | |
| "eval_runtime": 2.3345, | |
| "eval_samples_per_second": 307.127, | |
| "eval_steps_per_second": 9.852, | |
| "eval_wer": 0.09793126459793126, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 67.92, | |
| "learning_rate": 5.83659325005591e-06, | |
| "loss": 2.6997, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 67.92, | |
| "eval_loss": 3.2009031772613525, | |
| "eval_runtime": 2.2428, | |
| "eval_samples_per_second": 319.684, | |
| "eval_steps_per_second": 10.255, | |
| "eval_wer": 0.09793126459793126, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 68.87, | |
| "learning_rate": 5.53949547194521e-06, | |
| "loss": 2.5935, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 68.87, | |
| "eval_loss": 3.2141449451446533, | |
| "eval_runtime": 2.306, | |
| "eval_samples_per_second": 310.924, | |
| "eval_steps_per_second": 9.974, | |
| "eval_wer": 0.09784784784784785, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 69.81, | |
| "learning_rate": 5.247234378047524e-06, | |
| "loss": 2.6115, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 69.81, | |
| "eval_loss": 3.1760427951812744, | |
| "eval_runtime": 2.254, | |
| "eval_samples_per_second": 318.096, | |
| "eval_steps_per_second": 10.204, | |
| "eval_wer": 0.09467801134467801, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 70.75, | |
| "learning_rate": 4.960126877495005e-06, | |
| "loss": 2.5713, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 70.75, | |
| "eval_loss": 3.1937472820281982, | |
| "eval_runtime": 2.3527, | |
| "eval_samples_per_second": 304.754, | |
| "eval_steps_per_second": 9.776, | |
| "eval_wer": 0.09768101434768102, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 71.7, | |
| "learning_rate": 4.67848429119466e-06, | |
| "loss": 2.6647, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 71.7, | |
| "eval_loss": 3.162875175476074, | |
| "eval_runtime": 2.3463, | |
| "eval_samples_per_second": 305.584, | |
| "eval_steps_per_second": 9.803, | |
| "eval_wer": 0.0985985985985986, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 72.64, | |
| "learning_rate": 4.402612014251967e-06, | |
| "loss": 2.4878, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 72.64, | |
| "eval_loss": 3.1674914360046387, | |
| "eval_runtime": 2.32, | |
| "eval_samples_per_second": 309.054, | |
| "eval_steps_per_second": 9.914, | |
| "eval_wer": 0.09517851184517852, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 73.58, | |
| "learning_rate": 4.132809184820095e-06, | |
| "loss": 2.4761, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 73.58, | |
| "eval_loss": 3.1950597763061523, | |
| "eval_runtime": 2.2542, | |
| "eval_samples_per_second": 318.073, | |
| "eval_steps_per_second": 10.203, | |
| "eval_wer": 0.09759759759759759, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 74.53, | |
| "learning_rate": 3.869368359733711e-06, | |
| "loss": 2.3124, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 74.53, | |
| "eval_loss": 3.162883758544922, | |
| "eval_runtime": 2.3024, | |
| "eval_samples_per_second": 311.42, | |
| "eval_steps_per_second": 9.99, | |
| "eval_wer": 0.09542876209542876, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 75.47, | |
| "learning_rate": 3.6125751972791635e-06, | |
| "loss": 2.5718, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 75.47, | |
| "eval_loss": 3.1576590538024902, | |
| "eval_runtime": 2.2697, | |
| "eval_samples_per_second": 315.908, | |
| "eval_steps_per_second": 10.134, | |
| "eval_wer": 0.09784784784784785, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 76.42, | |
| "learning_rate": 3.3627081474450273e-06, | |
| "loss": 2.4606, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 76.42, | |
| "eval_loss": 3.1632273197174072, | |
| "eval_runtime": 2.4587, | |
| "eval_samples_per_second": 291.613, | |
| "eval_steps_per_second": 9.354, | |
| "eval_wer": 0.09734734734734735, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 77.36, | |
| "learning_rate": 3.120038149988832e-06, | |
| "loss": 2.5313, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 77.36, | |
| "eval_loss": 3.1841392517089844, | |
| "eval_runtime": 2.3574, | |
| "eval_samples_per_second": 304.154, | |
| "eval_steps_per_second": 9.757, | |
| "eval_wer": 0.09884884884884886, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 78.3, | |
| "learning_rate": 2.889457763733774e-06, | |
| "loss": 2.5124, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 78.3, | |
| "eval_loss": 3.1893506050109863, | |
| "eval_runtime": 2.2547, | |
| "eval_samples_per_second": 318.009, | |
| "eval_steps_per_second": 10.201, | |
| "eval_wer": 0.09868201534868201, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 79.25, | |
| "learning_rate": 2.6618064384144925e-06, | |
| "loss": 2.3324, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 79.25, | |
| "eval_loss": 3.1718995571136475, | |
| "eval_runtime": 2.3078, | |
| "eval_samples_per_second": 310.686, | |
| "eval_steps_per_second": 9.966, | |
| "eval_wer": 0.0965965965965966, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 80.19, | |
| "learning_rate": 2.44211217820481e-06, | |
| "loss": 2.4468, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 80.19, | |
| "eval_loss": 3.175983190536499, | |
| "eval_runtime": 2.483, | |
| "eval_samples_per_second": 288.768, | |
| "eval_steps_per_second": 9.263, | |
| "eval_wer": 0.09642976309642977, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 81.13, | |
| "learning_rate": 2.2306132054298847e-06, | |
| "loss": 2.4035, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 81.13, | |
| "eval_loss": 3.201446294784546, | |
| "eval_runtime": 2.2646, | |
| "eval_samples_per_second": 316.615, | |
| "eval_steps_per_second": 10.156, | |
| "eval_wer": 0.09826493159826494, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 82.08, | |
| "learning_rate": 2.027538855972291e-06, | |
| "loss": 2.3834, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 82.08, | |
| "eval_loss": 3.1823039054870605, | |
| "eval_runtime": 2.3457, | |
| "eval_samples_per_second": 305.661, | |
| "eval_steps_per_second": 9.805, | |
| "eval_wer": 0.0965965965965966, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 83.02, | |
| "learning_rate": 1.8331093305949532e-06, | |
| "loss": 2.3655, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 83.02, | |
| "eval_loss": 3.1758322715759277, | |
| "eval_runtime": 2.3111, | |
| "eval_samples_per_second": 310.239, | |
| "eval_steps_per_second": 9.952, | |
| "eval_wer": 0.09476142809476143, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 83.96, | |
| "learning_rate": 1.647535456169591e-06, | |
| "loss": 2.3525, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 83.96, | |
| "eval_loss": 3.192059278488159, | |
| "eval_runtime": 2.397, | |
| "eval_samples_per_second": 299.127, | |
| "eval_steps_per_second": 9.595, | |
| "eval_wer": 0.09801468134801468, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 84.91, | |
| "learning_rate": 1.4710184570696184e-06, | |
| "loss": 2.4428, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 84.91, | |
| "eval_loss": 3.1989715099334717, | |
| "eval_runtime": 2.305, | |
| "eval_samples_per_second": 311.067, | |
| "eval_steps_per_second": 9.978, | |
| "eval_wer": 0.09701368034701369, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 85.85, | |
| "learning_rate": 1.3037497369753871e-06, | |
| "loss": 2.3276, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 85.85, | |
| "eval_loss": 3.1906816959381104, | |
| "eval_runtime": 2.3224, | |
| "eval_samples_per_second": 308.734, | |
| "eval_steps_per_second": 9.904, | |
| "eval_wer": 0.09843176509843177, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 86.79, | |
| "learning_rate": 1.1459106713283286e-06, | |
| "loss": 2.4423, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 86.79, | |
| "eval_loss": 3.189251661300659, | |
| "eval_runtime": 2.2565, | |
| "eval_samples_per_second": 317.748, | |
| "eval_steps_per_second": 10.193, | |
| "eval_wer": 0.09768101434768102, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 87.74, | |
| "learning_rate": 9.976724106591128e-07, | |
| "loss": 2.3457, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 87.74, | |
| "eval_loss": 3.2001450061798096, | |
| "eval_runtime": 2.4103, | |
| "eval_samples_per_second": 297.477, | |
| "eval_steps_per_second": 9.542, | |
| "eval_wer": 0.10051718385051718, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 88.68, | |
| "learning_rate": 8.591956950030067e-07, | |
| "loss": 2.4146, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 88.68, | |
| "eval_loss": 3.1883089542388916, | |
| "eval_runtime": 2.2661, | |
| "eval_samples_per_second": 316.404, | |
| "eval_steps_per_second": 10.15, | |
| "eval_wer": 0.09851518184851518, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 89.62, | |
| "learning_rate": 7.306306796037188e-07, | |
| "loss": 2.3415, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 89.62, | |
| "eval_loss": 3.193378210067749, | |
| "eval_runtime": 2.3137, | |
| "eval_samples_per_second": 309.9, | |
| "eval_steps_per_second": 9.941, | |
| "eval_wer": 0.09968301634968302, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 90.57, | |
| "learning_rate": 6.121167720947174e-07, | |
| "loss": 2.2057, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 90.57, | |
| "eval_loss": 3.193852663040161, | |
| "eval_runtime": 2.2411, | |
| "eval_samples_per_second": 319.931, | |
| "eval_steps_per_second": 10.263, | |
| "eval_wer": 0.09951618284951619, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 91.51, | |
| "learning_rate": 5.037824813345571e-07, | |
| "loss": 2.5141, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 91.51, | |
| "eval_loss": 3.1944401264190674, | |
| "eval_runtime": 2.2989, | |
| "eval_samples_per_second": 311.892, | |
| "eval_steps_per_second": 10.005, | |
| "eval_wer": 0.1006006006006006, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 92.45, | |
| "learning_rate": 4.057452780601334e-07, | |
| "loss": 2.175, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 92.45, | |
| "eval_loss": 3.1808109283447266, | |
| "eval_runtime": 2.257, | |
| "eval_samples_per_second": 317.681, | |
| "eval_steps_per_second": 10.191, | |
| "eval_wer": 0.0985985985985986, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 93.4, | |
| "learning_rate": 3.1811146750898025e-07, | |
| "loss": 2.4668, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 93.4, | |
| "eval_loss": 3.1884679794311523, | |
| "eval_runtime": 2.3016, | |
| "eval_samples_per_second": 311.519, | |
| "eval_steps_per_second": 9.993, | |
| "eval_wer": 0.09943276609943276, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 94.34, | |
| "learning_rate": 2.4097607414869995e-07, | |
| "loss": 2.2732, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 94.34, | |
| "eval_loss": 3.1876654624938965, | |
| "eval_runtime": 2.3135, | |
| "eval_samples_per_second": 309.92, | |
| "eval_steps_per_second": 9.942, | |
| "eval_wer": 0.09984984984984985, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 95.28, | |
| "learning_rate": 1.7442273863854553e-07, | |
| "loss": 2.2636, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 95.28, | |
| "eval_loss": 3.187673568725586, | |
| "eval_runtime": 2.2902, | |
| "eval_samples_per_second": 313.071, | |
| "eval_steps_per_second": 10.043, | |
| "eval_wer": 0.09893226559893227, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 96.23, | |
| "learning_rate": 1.185236271348722e-07, | |
| "loss": 2.3504, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 96.23, | |
| "eval_loss": 3.1904258728027344, | |
| "eval_runtime": 2.2424, | |
| "eval_samples_per_second": 319.748, | |
| "eval_steps_per_second": 10.257, | |
| "eval_wer": 0.10001668335001668, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 97.17, | |
| "learning_rate": 7.33393530387927e-08, | |
| "loss": 2.2721, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 97.17, | |
| "eval_loss": 3.1916568279266357, | |
| "eval_runtime": 2.3214, | |
| "eval_samples_per_second": 308.871, | |
| "eval_steps_per_second": 9.908, | |
| "eval_wer": 0.10051718385051718, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 98.11, | |
| "learning_rate": 3.8918911270908745e-08, | |
| "loss": 2.4014, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 98.11, | |
| "eval_loss": 3.1922268867492676, | |
| "eval_runtime": 2.3615, | |
| "eval_samples_per_second": 303.621, | |
| "eval_steps_per_second": 9.74, | |
| "eval_wer": 0.10026693360026694, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 99.06, | |
| "learning_rate": 1.5299625144370444e-08, | |
| "loss": 2.3263, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 99.06, | |
| "eval_loss": 3.189690351486206, | |
| "eval_runtime": 2.305, | |
| "eval_samples_per_second": 311.061, | |
| "eval_steps_per_second": 9.978, | |
| "eval_wer": 0.09984984984984985, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 2.6567448613734612e-09, | |
| "loss": 2.3731, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_loss": 3.1906578540802, | |
| "eval_runtime": 2.2424, | |
| "eval_samples_per_second": 319.749, | |
| "eval_steps_per_second": 10.257, | |
| "eval_wer": 0.1001001001001001, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 5300, | |
| "total_flos": 8.033701259431117e+16, | |
| "train_loss": 39.75085498593888, | |
| "train_runtime": 4049.5415, | |
| "train_samples_per_second": 83.565, | |
| "train_steps_per_second": 1.309 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 5300, | |
| "num_train_epochs": 100, | |
| "save_steps": 50, | |
| "total_flos": 8.033701259431117e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |