whisper-tiny-nob / trainer_state.json
pere's picture
End of training
dc1e2ff
raw
history blame
236 kB
{
"best_metric": 24.878197320341048,
"best_model_checkpoint": "./checkpoint-98000",
"epoch": 4.00848,
"global_step": 100000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 1.5000000000000002e-07,
"loss": 2.936,
"step": 50
},
{
"epoch": 0.0,
"learning_rate": 3.0000000000000004e-07,
"loss": 2.8713,
"step": 100
},
{
"epoch": 0.0,
"learning_rate": 4.5e-07,
"loss": 2.7643,
"step": 150
},
{
"epoch": 0.0,
"learning_rate": 6.000000000000001e-07,
"loss": 2.6478,
"step": 200
},
{
"epoch": 0.0,
"learning_rate": 7.5e-07,
"loss": 2.5266,
"step": 250
},
{
"epoch": 0.0,
"learning_rate": 9e-07,
"loss": 2.4391,
"step": 300
},
{
"epoch": 0.0,
"learning_rate": 1.05e-06,
"loss": 2.3618,
"step": 350
},
{
"epoch": 0.0,
"learning_rate": 1.2000000000000002e-06,
"loss": 2.2763,
"step": 400
},
{
"epoch": 0.0,
"learning_rate": 1.35e-06,
"loss": 2.2461,
"step": 450
},
{
"epoch": 0.01,
"learning_rate": 1.5e-06,
"loss": 2.1815,
"step": 500
},
{
"epoch": 0.01,
"learning_rate": 1.65e-06,
"loss": 2.1294,
"step": 550
},
{
"epoch": 0.01,
"learning_rate": 1.8e-06,
"loss": 2.0958,
"step": 600
},
{
"epoch": 0.01,
"learning_rate": 1.95e-06,
"loss": 2.0656,
"step": 650
},
{
"epoch": 0.01,
"learning_rate": 2.1e-06,
"loss": 2.0392,
"step": 700
},
{
"epoch": 0.01,
"learning_rate": 2.25e-06,
"loss": 2.0037,
"step": 750
},
{
"epoch": 0.01,
"learning_rate": 2.4000000000000003e-06,
"loss": 1.9791,
"step": 800
},
{
"epoch": 0.01,
"learning_rate": 2.55e-06,
"loss": 1.9636,
"step": 850
},
{
"epoch": 0.01,
"learning_rate": 2.7e-06,
"loss": 1.9405,
"step": 900
},
{
"epoch": 0.01,
"learning_rate": 2.85e-06,
"loss": 1.9216,
"step": 950
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.8819,
"step": 1000
},
{
"epoch": 0.01,
"eval_loss": 1.1868911981582642,
"eval_runtime": 54.8495,
"eval_samples_per_second": 2.935,
"eval_steps_per_second": 0.055,
"eval_wer": 61.96711327649208,
"step": 1000
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.8776,
"step": 1050
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.8508,
"step": 1100
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.8361,
"step": 1150
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.8191,
"step": 1200
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.7992,
"step": 1250
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.7928,
"step": 1300
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.7808,
"step": 1350
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.7558,
"step": 1400
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.7655,
"step": 1450
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.7327,
"step": 1500
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.7395,
"step": 1550
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.7301,
"step": 1600
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.7016,
"step": 1650
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.6962,
"step": 1700
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.6992,
"step": 1750
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.6888,
"step": 1800
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.6646,
"step": 1850
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.6713,
"step": 1900
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.6562,
"step": 1950
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.6425,
"step": 2000
},
{
"epoch": 0.02,
"eval_loss": 0.9990558624267578,
"eval_runtime": 38.5229,
"eval_samples_per_second": 4.179,
"eval_steps_per_second": 0.078,
"eval_wer": 53.65408038976858,
"step": 2000
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.6376,
"step": 2050
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.6457,
"step": 2100
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.6244,
"step": 2150
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.6113,
"step": 2200
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.6096,
"step": 2250
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.6302,
"step": 2300
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.6078,
"step": 2350
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.6064,
"step": 2400
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.5884,
"step": 2450
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5819,
"step": 2500
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5784,
"step": 2550
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5648,
"step": 2600
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5758,
"step": 2650
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5627,
"step": 2700
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5573,
"step": 2750
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5594,
"step": 2800
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.55,
"step": 2850
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5583,
"step": 2900
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5416,
"step": 2950
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.548,
"step": 3000
},
{
"epoch": 0.03,
"eval_loss": 0.9147223234176636,
"eval_runtime": 37.9927,
"eval_samples_per_second": 4.238,
"eval_steps_per_second": 0.079,
"eval_wer": 50.21315468940317,
"step": 3000
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5347,
"step": 3050
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5435,
"step": 3100
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5246,
"step": 3150
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5326,
"step": 3200
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5094,
"step": 3250
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5061,
"step": 3300
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5139,
"step": 3350
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5055,
"step": 3400
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.5109,
"step": 3450
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4999,
"step": 3500
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.5015,
"step": 3550
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4853,
"step": 3600
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4836,
"step": 3650
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4874,
"step": 3700
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4758,
"step": 3750
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.462,
"step": 3800
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4813,
"step": 3850
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4836,
"step": 3900
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4692,
"step": 3950
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4636,
"step": 4000
},
{
"epoch": 0.04,
"eval_loss": 0.8605496287345886,
"eval_runtime": 40.5349,
"eval_samples_per_second": 3.972,
"eval_steps_per_second": 0.074,
"eval_wer": 47.07673568818514,
"step": 4000
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4715,
"step": 4050
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.477,
"step": 4100
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4539,
"step": 4150
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4555,
"step": 4200
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4423,
"step": 4250
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4506,
"step": 4300
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4404,
"step": 4350
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4455,
"step": 4400
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4463,
"step": 4450
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.4324,
"step": 4500
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.4317,
"step": 4550
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.4169,
"step": 4600
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.4282,
"step": 4650
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.4487,
"step": 4700
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.4348,
"step": 4750
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.4173,
"step": 4800
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.4247,
"step": 4850
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.4088,
"step": 4900
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.4198,
"step": 4950
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.4113,
"step": 5000
},
{
"epoch": 0.05,
"eval_loss": 0.8252834677696228,
"eval_runtime": 38.0695,
"eval_samples_per_second": 4.229,
"eval_steps_per_second": 0.079,
"eval_wer": 45.73690621193666,
"step": 5000
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 1.4062,
"step": 5050
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 1.3963,
"step": 5100
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 1.4073,
"step": 5150
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 1.3978,
"step": 5200
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 1.3838,
"step": 5250
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 1.3836,
"step": 5300
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 1.3974,
"step": 5350
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 1.3793,
"step": 5400
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 1.4009,
"step": 5450
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3867,
"step": 5500
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3756,
"step": 5550
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3735,
"step": 5600
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3735,
"step": 5650
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3719,
"step": 5700
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3661,
"step": 5750
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3646,
"step": 5800
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.373,
"step": 5850
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3669,
"step": 5900
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3658,
"step": 5950
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3484,
"step": 6000
},
{
"epoch": 0.01,
"eval_loss": 0.7946101427078247,
"eval_runtime": 39.2317,
"eval_samples_per_second": 4.104,
"eval_steps_per_second": 0.076,
"eval_wer": 43.4531059683313,
"step": 6000
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3619,
"step": 6050
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.355,
"step": 6100
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3529,
"step": 6150
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3469,
"step": 6200
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3411,
"step": 6250
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3491,
"step": 6300
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.352,
"step": 6350
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3355,
"step": 6400
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3545,
"step": 6450
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 1.3312,
"step": 6500
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3494,
"step": 6550
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3443,
"step": 6600
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3241,
"step": 6650
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3295,
"step": 6700
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3391,
"step": 6750
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3323,
"step": 6800
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3168,
"step": 6850
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3301,
"step": 6900
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3225,
"step": 6950
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3127,
"step": 7000
},
{
"epoch": 0.02,
"eval_loss": 0.7740240097045898,
"eval_runtime": 39.5487,
"eval_samples_per_second": 4.071,
"eval_steps_per_second": 0.076,
"eval_wer": 42.265529841656516,
"step": 7000
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.314,
"step": 7050
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3271,
"step": 7100
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3121,
"step": 7150
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3037,
"step": 7200
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3075,
"step": 7250
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.328,
"step": 7300
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3154,
"step": 7350
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3151,
"step": 7400
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 1.3032,
"step": 7450
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2986,
"step": 7500
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2994,
"step": 7550
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2927,
"step": 7600
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.305,
"step": 7650
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2958,
"step": 7700
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2945,
"step": 7750
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2989,
"step": 7800
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2951,
"step": 7850
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.305,
"step": 7900
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2923,
"step": 7950
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2994,
"step": 8000
},
{
"epoch": 0.03,
"eval_loss": 0.7550716996192932,
"eval_runtime": 39.5194,
"eval_samples_per_second": 4.074,
"eval_steps_per_second": 0.076,
"eval_wer": 40.895249695493305,
"step": 8000
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2896,
"step": 8050
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.3027,
"step": 8100
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2868,
"step": 8150
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2963,
"step": 8200
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2774,
"step": 8250
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2754,
"step": 8300
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2844,
"step": 8350
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2806,
"step": 8400
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 1.2889,
"step": 8450
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2804,
"step": 8500
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2835,
"step": 8550
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2706,
"step": 8600
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2696,
"step": 8650
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2776,
"step": 8700
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2686,
"step": 8750
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2566,
"step": 8800
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2762,
"step": 8850
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2805,
"step": 8900
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2679,
"step": 8950
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.265,
"step": 9000
},
{
"epoch": 0.04,
"eval_loss": 0.737849235534668,
"eval_runtime": 40.6621,
"eval_samples_per_second": 3.959,
"eval_steps_per_second": 0.074,
"eval_wer": 39.859926918392205,
"step": 9000
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2748,
"step": 9050
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2816,
"step": 9100
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2618,
"step": 9150
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2648,
"step": 9200
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2532,
"step": 9250
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2633,
"step": 9300
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2545,
"step": 9350
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2625,
"step": 9400
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2641,
"step": 9450
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 1.2514,
"step": 9500
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2529,
"step": 9550
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2397,
"step": 9600
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2534,
"step": 9650
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.273,
"step": 9700
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.263,
"step": 9750
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2452,
"step": 9800
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2544,
"step": 9850
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2393,
"step": 9900
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2516,
"step": 9950
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2458,
"step": 10000
},
{
"epoch": 0.05,
"eval_loss": 0.7256616353988647,
"eval_runtime": 39.9578,
"eval_samples_per_second": 4.029,
"eval_steps_per_second": 0.075,
"eval_wer": 39.89037758830694,
"step": 10000
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2579,
"step": 10050
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2544,
"step": 10100
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2491,
"step": 10150
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2535,
"step": 10200
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2515,
"step": 10250
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2431,
"step": 10300
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2503,
"step": 10350
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.256,
"step": 10400
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 1.2445,
"step": 10450
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2336,
"step": 10500
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.237,
"step": 10550
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.235,
"step": 10600
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2435,
"step": 10650
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2456,
"step": 10700
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2269,
"step": 10750
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2289,
"step": 10800
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2379,
"step": 10850
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2261,
"step": 10900
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2363,
"step": 10950
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2257,
"step": 11000
},
{
"epoch": 0.06,
"eval_loss": 0.7114033102989197,
"eval_runtime": 39.5013,
"eval_samples_per_second": 4.076,
"eval_steps_per_second": 0.076,
"eval_wer": 39.79902557856273,
"step": 11000
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2315,
"step": 11050
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.229,
"step": 11100
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2291,
"step": 11150
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2311,
"step": 11200
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2186,
"step": 11250
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2326,
"step": 11300
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2277,
"step": 11350
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2311,
"step": 11400
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 1.2213,
"step": 11450
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2163,
"step": 11500
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2171,
"step": 11550
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2288,
"step": 11600
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2184,
"step": 11650
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2118,
"step": 11700
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2137,
"step": 11750
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.211,
"step": 11800
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2264,
"step": 11850
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2133,
"step": 11900
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2104,
"step": 11950
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2126,
"step": 12000
},
{
"epoch": 0.07,
"eval_loss": 0.6972322463989258,
"eval_runtime": 40.3496,
"eval_samples_per_second": 3.99,
"eval_steps_per_second": 0.074,
"eval_wer": 37.880633373934224,
"step": 12000
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2101,
"step": 12050
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2027,
"step": 12100
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.207,
"step": 12150
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2193,
"step": 12200
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2064,
"step": 12250
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2109,
"step": 12300
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2048,
"step": 12350
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.195,
"step": 12400
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.2037,
"step": 12450
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 1.1998,
"step": 12500
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 1.1935,
"step": 12550
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 1.1955,
"step": 12600
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 1.1994,
"step": 12650
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 1.2118,
"step": 12700
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 1.2017,
"step": 12750
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 1.2088,
"step": 12800
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 1.1909,
"step": 12850
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 1.2045,
"step": 12900
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 1.2031,
"step": 12950
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 1.1971,
"step": 13000
},
{
"epoch": 0.08,
"eval_loss": 0.6871449947357178,
"eval_runtime": 39.2495,
"eval_samples_per_second": 4.102,
"eval_steps_per_second": 0.076,
"eval_wer": 37.3020706455542,
"step": 13000
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 1.2081,
"step": 13050
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 1.1995,
"step": 13100
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 1.1953,
"step": 13150
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 1.1944,
"step": 13200
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 1.1872,
"step": 13250
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 1.177,
"step": 13300
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 1.1792,
"step": 13350
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 1.1717,
"step": 13400
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 1.1523,
"step": 13450
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 1.1751,
"step": 13500
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 1.1815,
"step": 13550
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 1.183,
"step": 13600
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 1.17,
"step": 13650
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 1.1812,
"step": 13700
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 1.1798,
"step": 13750
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1626,
"step": 13800
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1687,
"step": 13850
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1676,
"step": 13900
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1748,
"step": 13950
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1786,
"step": 14000
},
{
"epoch": 1.01,
"eval_loss": 0.6785603761672974,
"eval_runtime": 39.6356,
"eval_samples_per_second": 4.062,
"eval_steps_per_second": 0.076,
"eval_wer": 37.42387332521315,
"step": 14000
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.172,
"step": 14050
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1752,
"step": 14100
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1765,
"step": 14150
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1586,
"step": 14200
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1578,
"step": 14250
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1555,
"step": 14300
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1659,
"step": 14350
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1701,
"step": 14400
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1687,
"step": 14450
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1804,
"step": 14500
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1636,
"step": 14550
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1559,
"step": 14600
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1555,
"step": 14650
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1563,
"step": 14700
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 1.1508,
"step": 14750
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1454,
"step": 14800
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1555,
"step": 14850
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1627,
"step": 14900
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1578,
"step": 14950
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1486,
"step": 15000
},
{
"epoch": 1.02,
"eval_loss": 0.6702972054481506,
"eval_runtime": 43.4585,
"eval_samples_per_second": 3.705,
"eval_steps_per_second": 0.069,
"eval_wer": 36.99756394640682,
"step": 15000
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1655,
"step": 15050
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1649,
"step": 15100
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1609,
"step": 15150
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1516,
"step": 15200
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1635,
"step": 15250
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1528,
"step": 15300
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1688,
"step": 15350
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1508,
"step": 15400
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1511,
"step": 15450
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1354,
"step": 15500
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1562,
"step": 15550
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1523,
"step": 15600
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1469,
"step": 15650
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1458,
"step": 15700
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 1.1455,
"step": 15750
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1444,
"step": 15800
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1582,
"step": 15850
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1447,
"step": 15900
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1494,
"step": 15950
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1505,
"step": 16000
},
{
"epoch": 1.03,
"eval_loss": 0.6647058129310608,
"eval_runtime": 39.1012,
"eval_samples_per_second": 4.118,
"eval_steps_per_second": 0.077,
"eval_wer": 36.35809987819732,
"step": 16000
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.148,
"step": 16050
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.151,
"step": 16100
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.149,
"step": 16150
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1413,
"step": 16200
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1398,
"step": 16250
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1406,
"step": 16300
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1335,
"step": 16350
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1326,
"step": 16400
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1425,
"step": 16450
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.148,
"step": 16500
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1368,
"step": 16550
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1308,
"step": 16600
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1298,
"step": 16650
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1352,
"step": 16700
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 1.1327,
"step": 16750
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1409,
"step": 16800
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1263,
"step": 16850
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1304,
"step": 16900
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1298,
"step": 16950
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1238,
"step": 17000
},
{
"epoch": 1.04,
"eval_loss": 0.6559091210365295,
"eval_runtime": 39.2532,
"eval_samples_per_second": 4.102,
"eval_steps_per_second": 0.076,
"eval_wer": 36.38855054811206,
"step": 17000
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1387,
"step": 17050
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1357,
"step": 17100
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1308,
"step": 17150
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1327,
"step": 17200
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1278,
"step": 17250
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1384,
"step": 17300
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1433,
"step": 17350
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1354,
"step": 17400
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1356,
"step": 17450
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1271,
"step": 17500
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1263,
"step": 17550
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1367,
"step": 17600
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1222,
"step": 17650
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1357,
"step": 17700
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 1.1127,
"step": 17750
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1303,
"step": 17800
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1174,
"step": 17850
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1407,
"step": 17900
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1321,
"step": 17950
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1184,
"step": 18000
},
{
"epoch": 1.05,
"eval_loss": 0.6509166955947876,
"eval_runtime": 39.722,
"eval_samples_per_second": 4.053,
"eval_steps_per_second": 0.076,
"eval_wer": 36.510353227771006,
"step": 18000
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1373,
"step": 18050
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1208,
"step": 18100
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1138,
"step": 18150
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1173,
"step": 18200
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1235,
"step": 18250
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1106,
"step": 18300
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1157,
"step": 18350
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1321,
"step": 18400
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1391,
"step": 18450
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.112,
"step": 18500
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1163,
"step": 18550
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1226,
"step": 18600
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1053,
"step": 18650
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1184,
"step": 18700
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 1.1134,
"step": 18750
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.1112,
"step": 18800
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.1194,
"step": 18850
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.1184,
"step": 18900
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.1182,
"step": 18950
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.115,
"step": 19000
},
{
"epoch": 1.06,
"eval_loss": 0.6451593041419983,
"eval_runtime": 41.32,
"eval_samples_per_second": 3.896,
"eval_steps_per_second": 0.073,
"eval_wer": 35.99269183922046,
"step": 19000
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.121,
"step": 19050
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.1133,
"step": 19100
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.1041,
"step": 19150
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.1116,
"step": 19200
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.112,
"step": 19250
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.112,
"step": 19300
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.1122,
"step": 19350
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.1059,
"step": 19400
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.1087,
"step": 19450
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.1146,
"step": 19500
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.1068,
"step": 19550
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.1058,
"step": 19600
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.112,
"step": 19650
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.0969,
"step": 19700
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 1.1134,
"step": 19750
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.0999,
"step": 19800
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.1006,
"step": 19850
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.1019,
"step": 19900
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.1105,
"step": 19950
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.1013,
"step": 20000
},
{
"epoch": 1.07,
"eval_loss": 0.6382384896278381,
"eval_runtime": 40.3978,
"eval_samples_per_second": 3.985,
"eval_steps_per_second": 0.074,
"eval_wer": 34.50060901339829,
"step": 20000
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.121,
"step": 20050
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.1122,
"step": 20100
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.1039,
"step": 20150
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.1013,
"step": 20200
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.1097,
"step": 20250
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.1002,
"step": 20300
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.0965,
"step": 20350
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.0903,
"step": 20400
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.1016,
"step": 20450
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.0923,
"step": 20500
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.0983,
"step": 20550
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.1011,
"step": 20600
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.0963,
"step": 20650
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.0952,
"step": 20700
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 1.0972,
"step": 20750
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.1087,
"step": 20800
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.0958,
"step": 20850
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.0902,
"step": 20900
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.0958,
"step": 20950
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.0969,
"step": 21000
},
{
"epoch": 1.08,
"eval_loss": 0.633127748966217,
"eval_runtime": 40.5043,
"eval_samples_per_second": 3.975,
"eval_steps_per_second": 0.074,
"eval_wer": 34.348355663824606,
"step": 21000
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.0988,
"step": 21050
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.0947,
"step": 21100
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.0964,
"step": 21150
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.0977,
"step": 21200
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.0986,
"step": 21250
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.1004,
"step": 21300
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.0951,
"step": 21350
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.086,
"step": 21400
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.0905,
"step": 21450
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.0967,
"step": 21500
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 1.079,
"step": 21550
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 1.0978,
"step": 21600
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 1.0779,
"step": 21650
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 1.0855,
"step": 21700
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 1.0775,
"step": 21750
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 1.068,
"step": 21800
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 1.0789,
"step": 21850
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 1.0748,
"step": 21900
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 1.075,
"step": 21950
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 1.0784,
"step": 22000
},
{
"epoch": 2.0,
"eval_loss": 0.6303825974464417,
"eval_runtime": 38.4993,
"eval_samples_per_second": 4.182,
"eval_steps_per_second": 0.078,
"eval_wer": 34.28745432399513,
"step": 22000
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 1.0746,
"step": 22050
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0761,
"step": 22100
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0793,
"step": 22150
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0814,
"step": 22200
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0801,
"step": 22250
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0791,
"step": 22300
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0718,
"step": 22350
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0855,
"step": 22400
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0949,
"step": 22450
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0627,
"step": 22500
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0873,
"step": 22550
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0698,
"step": 22600
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0667,
"step": 22650
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0735,
"step": 22700
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0821,
"step": 22750
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0794,
"step": 22800
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0646,
"step": 22850
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0684,
"step": 22900
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0673,
"step": 22950
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.0774,
"step": 23000
},
{
"epoch": 2.01,
"eval_loss": 0.6248754262924194,
"eval_runtime": 38.6299,
"eval_samples_per_second": 4.168,
"eval_steps_per_second": 0.078,
"eval_wer": 34.104750304506695,
"step": 23000
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 1.065,
"step": 23050
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0757,
"step": 23100
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0676,
"step": 23150
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0604,
"step": 23200
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.071,
"step": 23250
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0778,
"step": 23300
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.063,
"step": 23350
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0705,
"step": 23400
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0639,
"step": 23450
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.057,
"step": 23500
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0617,
"step": 23550
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0623,
"step": 23600
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0778,
"step": 23650
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0687,
"step": 23700
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0626,
"step": 23750
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0781,
"step": 23800
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0579,
"step": 23850
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0624,
"step": 23900
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0719,
"step": 23950
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0719,
"step": 24000
},
{
"epoch": 2.02,
"eval_loss": 0.6194329857826233,
"eval_runtime": 39.0738,
"eval_samples_per_second": 4.12,
"eval_steps_per_second": 0.077,
"eval_wer": 33.830694275274055,
"step": 24000
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 1.0779,
"step": 24050
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0651,
"step": 24100
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0654,
"step": 24150
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0647,
"step": 24200
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0693,
"step": 24250
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0682,
"step": 24300
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0733,
"step": 24350
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0597,
"step": 24400
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0737,
"step": 24450
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0547,
"step": 24500
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.063,
"step": 24550
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0582,
"step": 24600
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.06,
"step": 24650
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0713,
"step": 24700
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0699,
"step": 24750
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0552,
"step": 24800
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0566,
"step": 24850
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.048,
"step": 24900
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0529,
"step": 24950
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0638,
"step": 25000
},
{
"epoch": 2.03,
"eval_loss": 0.6157576441764832,
"eval_runtime": 39.7564,
"eval_samples_per_second": 4.05,
"eval_steps_per_second": 0.075,
"eval_wer": 32.97807551766139,
"step": 25000
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 1.0555,
"step": 25050
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0485,
"step": 25100
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0554,
"step": 25150
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.061,
"step": 25200
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0582,
"step": 25250
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0426,
"step": 25300
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0551,
"step": 25350
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0577,
"step": 25400
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0811,
"step": 25450
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0541,
"step": 25500
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0446,
"step": 25550
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0642,
"step": 25600
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0492,
"step": 25650
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0469,
"step": 25700
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.052,
"step": 25750
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0534,
"step": 25800
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0457,
"step": 25850
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0459,
"step": 25900
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0684,
"step": 25950
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0592,
"step": 26000
},
{
"epoch": 2.04,
"eval_loss": 0.610471785068512,
"eval_runtime": 40.4875,
"eval_samples_per_second": 3.977,
"eval_steps_per_second": 0.074,
"eval_wer": 32.64311814859927,
"step": 26000
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 1.0649,
"step": 26050
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0508,
"step": 26100
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0518,
"step": 26150
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0587,
"step": 26200
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.05,
"step": 26250
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0449,
"step": 26300
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0514,
"step": 26350
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0542,
"step": 26400
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0587,
"step": 26450
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0566,
"step": 26500
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0393,
"step": 26550
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0537,
"step": 26600
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0457,
"step": 26650
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0475,
"step": 26700
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0591,
"step": 26750
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0476,
"step": 26800
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0537,
"step": 26850
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0414,
"step": 26900
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0596,
"step": 26950
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.0493,
"step": 27000
},
{
"epoch": 2.05,
"eval_loss": 0.6040655970573425,
"eval_runtime": 38.168,
"eval_samples_per_second": 4.218,
"eval_steps_per_second": 0.079,
"eval_wer": 32.734470158343484,
"step": 27000
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 1.048,
"step": 27050
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0542,
"step": 27100
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0474,
"step": 27150
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0424,
"step": 27200
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0337,
"step": 27250
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0424,
"step": 27300
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0342,
"step": 27350
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0401,
"step": 27400
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0388,
"step": 27450
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0323,
"step": 27500
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0361,
"step": 27550
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0322,
"step": 27600
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0474,
"step": 27650
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.05,
"step": 27700
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0501,
"step": 27750
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0425,
"step": 27800
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0421,
"step": 27850
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0453,
"step": 27900
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0352,
"step": 27950
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.047,
"step": 28000
},
{
"epoch": 2.06,
"eval_loss": 0.6039990186691284,
"eval_runtime": 39.0034,
"eval_samples_per_second": 4.128,
"eval_steps_per_second": 0.077,
"eval_wer": 32.76492082825822,
"step": 28000
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 1.0366,
"step": 28050
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0588,
"step": 28100
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0429,
"step": 28150
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0461,
"step": 28200
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0347,
"step": 28250
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.038,
"step": 28300
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0361,
"step": 28350
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0226,
"step": 28400
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0337,
"step": 28450
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0351,
"step": 28500
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0279,
"step": 28550
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0411,
"step": 28600
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0384,
"step": 28650
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0434,
"step": 28700
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0414,
"step": 28750
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0305,
"step": 28800
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0301,
"step": 28850
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0293,
"step": 28900
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0324,
"step": 28950
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0323,
"step": 29000
},
{
"epoch": 2.07,
"eval_loss": 0.5984179377555847,
"eval_runtime": 38.0029,
"eval_samples_per_second": 4.237,
"eval_steps_per_second": 0.079,
"eval_wer": 31.60779537149817,
"step": 29000
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 1.0444,
"step": 29050
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0328,
"step": 29100
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0334,
"step": 29150
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0332,
"step": 29200
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0297,
"step": 29250
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0441,
"step": 29300
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0319,
"step": 29350
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0454,
"step": 29400
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0286,
"step": 29450
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0227,
"step": 29500
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0299,
"step": 29550
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0317,
"step": 29600
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0227,
"step": 29650
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0374,
"step": 29700
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0352,
"step": 29750
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0176,
"step": 29800
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 1.0285,
"step": 29850
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 1.0243,
"step": 29900
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 1.0192,
"step": 29950
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 1.0189,
"step": 30000
},
{
"epoch": 3.0,
"eval_loss": 0.5957360863685608,
"eval_runtime": 30.8132,
"eval_samples_per_second": 5.225,
"eval_steps_per_second": 0.097,
"eval_wer": 31.303288672350792,
"step": 30000
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 1.0133,
"step": 30050
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 1.0144,
"step": 30100
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 1.0279,
"step": 30150
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 1.0148,
"step": 30200
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 1.0192,
"step": 30250
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 1.0246,
"step": 30300
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 1.026,
"step": 30350
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0173,
"step": 30400
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0273,
"step": 30450
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0306,
"step": 30500
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0115,
"step": 30550
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0172,
"step": 30600
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0068,
"step": 30650
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0066,
"step": 30700
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0204,
"step": 30750
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0206,
"step": 30800
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0158,
"step": 30850
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0117,
"step": 30900
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0096,
"step": 30950
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0078,
"step": 31000
},
{
"epoch": 3.01,
"eval_loss": 0.5924085378646851,
"eval_runtime": 31.4548,
"eval_samples_per_second": 5.118,
"eval_steps_per_second": 0.095,
"eval_wer": 31.425091352009744,
"step": 31000
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0176,
"step": 31050
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0132,
"step": 31100
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.02,
"step": 31150
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0144,
"step": 31200
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0031,
"step": 31250
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0067,
"step": 31300
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 1.0025,
"step": 31350
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0099,
"step": 31400
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0186,
"step": 31450
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0235,
"step": 31500
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.015,
"step": 31550
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0154,
"step": 31600
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0117,
"step": 31650
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0027,
"step": 31700
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0134,
"step": 31750
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0055,
"step": 31800
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0027,
"step": 31850
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0099,
"step": 31900
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0037,
"step": 31950
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0146,
"step": 32000
},
{
"epoch": 3.02,
"eval_loss": 0.594041645526886,
"eval_runtime": 32.4443,
"eval_samples_per_second": 4.962,
"eval_steps_per_second": 0.092,
"eval_wer": 31.303288672350792,
"step": 32000
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0069,
"step": 32050
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0082,
"step": 32100
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0154,
"step": 32150
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0227,
"step": 32200
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.9945,
"step": 32250
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0048,
"step": 32300
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 1.0101,
"step": 32350
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.9971,
"step": 32400
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.0,
"step": 32450
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.006,
"step": 32500
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.0083,
"step": 32550
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.006,
"step": 32600
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.0073,
"step": 32650
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.0048,
"step": 32700
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.0015,
"step": 32750
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.0101,
"step": 32800
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.01,
"step": 32850
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.0133,
"step": 32900
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.0069,
"step": 32950
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.0128,
"step": 33000
},
{
"epoch": 3.03,
"eval_loss": 0.5891727805137634,
"eval_runtime": 32.6258,
"eval_samples_per_second": 4.935,
"eval_steps_per_second": 0.092,
"eval_wer": 31.02923264311815,
"step": 33000
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.0075,
"step": 33050
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.0175,
"step": 33100
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.0083,
"step": 33150
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.9995,
"step": 33200
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.9996,
"step": 33250
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.0072,
"step": 33300
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 1.0196,
"step": 33350
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 1.0071,
"step": 33400
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 1.0073,
"step": 33450
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 1.012,
"step": 33500
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 1.0137,
"step": 33550
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.9993,
"step": 33600
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 1.0025,
"step": 33650
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 1.0055,
"step": 33700
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 1.0013,
"step": 33750
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 1.0068,
"step": 33800
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 1.0018,
"step": 33850
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.9998,
"step": 33900
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.9934,
"step": 33950
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 1.0025,
"step": 34000
},
{
"epoch": 3.04,
"eval_loss": 0.5873314738273621,
"eval_runtime": 32.6141,
"eval_samples_per_second": 4.937,
"eval_steps_per_second": 0.092,
"eval_wer": 31.181485992691837,
"step": 34000
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 1.0072,
"step": 34050
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.9934,
"step": 34100
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 1.0179,
"step": 34150
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.9992,
"step": 34200
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.9973,
"step": 34250
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 1.0184,
"step": 34300
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 1.0049,
"step": 34350
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 1.009,
"step": 34400
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 1.0099,
"step": 34450
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 1.0086,
"step": 34500
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 1.0038,
"step": 34550
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.9973,
"step": 34600
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 1.0189,
"step": 34650
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 1.0017,
"step": 34700
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 1.0013,
"step": 34750
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.9962,
"step": 34800
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 1.0062,
"step": 34850
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.9956,
"step": 34900
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.9908,
"step": 34950
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.999,
"step": 35000
},
{
"epoch": 3.05,
"eval_loss": 0.5838064551353455,
"eval_runtime": 29.7975,
"eval_samples_per_second": 5.403,
"eval_steps_per_second": 0.101,
"eval_wer": 30.633373934226555,
"step": 35000
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 1.0019,
"step": 35050
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 1.0049,
"step": 35100
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 1.0116,
"step": 35150
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 1.0041,
"step": 35200
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 1.0047,
"step": 35250
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 1.003,
"step": 35300
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.9984,
"step": 35350
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 1.0072,
"step": 35400
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.9937,
"step": 35450
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.9983,
"step": 35500
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 1.0007,
"step": 35550
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.998,
"step": 35600
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.9986,
"step": 35650
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 1.0009,
"step": 35700
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.9982,
"step": 35750
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 1.0009,
"step": 35800
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.9946,
"step": 35850
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 1.0014,
"step": 35900
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 1.0009,
"step": 35950
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 1.0045,
"step": 36000
},
{
"epoch": 3.06,
"eval_loss": 0.5799316167831421,
"eval_runtime": 35.1205,
"eval_samples_per_second": 4.584,
"eval_steps_per_second": 0.085,
"eval_wer": 30.420219244823386,
"step": 36000
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.9923,
"step": 36050
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.9973,
"step": 36100
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.9879,
"step": 36150
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.9857,
"step": 36200
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.9931,
"step": 36250
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.9926,
"step": 36300
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.9892,
"step": 36350
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.982,
"step": 36400
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.9962,
"step": 36450
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.9871,
"step": 36500
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.9868,
"step": 36550
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 1.0011,
"step": 36600
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.9873,
"step": 36650
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.9946,
"step": 36700
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 1.004,
"step": 36750
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.9963,
"step": 36800
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.9822,
"step": 36850
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.9949,
"step": 36900
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.9866,
"step": 36950
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 1.0005,
"step": 37000
},
{
"epoch": 3.07,
"eval_loss": 0.5770267248153687,
"eval_runtime": 32.6377,
"eval_samples_per_second": 4.933,
"eval_steps_per_second": 0.092,
"eval_wer": 30.176613885505482,
"step": 37000
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.9882,
"step": 37050
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.9887,
"step": 37100
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.9863,
"step": 37150
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.9874,
"step": 37200
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.996,
"step": 37250
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 1.0031,
"step": 37300
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.9861,
"step": 37350
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.9843,
"step": 37400
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.9873,
"step": 37450
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.9939,
"step": 37500
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.9777,
"step": 37550
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.9785,
"step": 37600
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 1.0,
"step": 37650
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.9911,
"step": 37700
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.9952,
"step": 37750
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.9814,
"step": 37800
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.985,
"step": 37850
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.9881,
"step": 37900
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.9882,
"step": 37950
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 1.0017,
"step": 38000
},
{
"epoch": 3.08,
"eval_loss": 0.5733200907707214,
"eval_runtime": 38.0152,
"eval_samples_per_second": 4.235,
"eval_steps_per_second": 0.079,
"eval_wer": 29.65895249695493,
"step": 38000
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.987,
"step": 38050
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.9886,
"step": 38100
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.9859,
"step": 38150
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.9752,
"step": 38200
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.9817,
"step": 38250
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.9751,
"step": 38300
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.9714,
"step": 38350
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.9681,
"step": 38400
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.9914,
"step": 38450
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.9775,
"step": 38500
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.9779,
"step": 38550
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.9697,
"step": 38600
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.979,
"step": 38650
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9747,
"step": 38700
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9777,
"step": 38750
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.973,
"step": 38800
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9783,
"step": 38850
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9719,
"step": 38900
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9732,
"step": 38950
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9878,
"step": 39000
},
{
"epoch": 4.01,
"eval_loss": 0.5744524002075195,
"eval_runtime": 41.2775,
"eval_samples_per_second": 3.9,
"eval_steps_per_second": 0.073,
"eval_wer": 30.267965895249695,
"step": 39000
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9827,
"step": 39050
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9796,
"step": 39100
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9726,
"step": 39150
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.984,
"step": 39200
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9739,
"step": 39250
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9885,
"step": 39300
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9724,
"step": 39350
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9701,
"step": 39400
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9749,
"step": 39450
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.965,
"step": 39500
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9805,
"step": 39550
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9829,
"step": 39600
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.9782,
"step": 39650
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9672,
"step": 39700
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9763,
"step": 39750
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9795,
"step": 39800
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9731,
"step": 39850
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9693,
"step": 39900
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9708,
"step": 39950
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9854,
"step": 40000
},
{
"epoch": 4.02,
"eval_loss": 0.5719765424728394,
"eval_runtime": 38.4572,
"eval_samples_per_second": 4.186,
"eval_steps_per_second": 0.078,
"eval_wer": 30.054811205846526,
"step": 40000
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9628,
"step": 40050
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9651,
"step": 40100
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9627,
"step": 40150
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9572,
"step": 40200
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9681,
"step": 40250
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9642,
"step": 40300
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9817,
"step": 40350
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9654,
"step": 40400
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9799,
"step": 40450
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.973,
"step": 40500
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9749,
"step": 40550
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9651,
"step": 40600
},
{
"epoch": 4.02,
"learning_rate": 3e-06,
"loss": 0.9686,
"step": 40650
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9808,
"step": 40700
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9778,
"step": 40750
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9736,
"step": 40800
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9767,
"step": 40850
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9866,
"step": 40900
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9708,
"step": 40950
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9624,
"step": 41000
},
{
"epoch": 4.03,
"eval_loss": 0.5703173279762268,
"eval_runtime": 39.8395,
"eval_samples_per_second": 4.041,
"eval_steps_per_second": 0.075,
"eval_wer": 29.598051157125454,
"step": 41000
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.975,
"step": 41050
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9664,
"step": 41100
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9583,
"step": 41150
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9644,
"step": 41200
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.967,
"step": 41250
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9632,
"step": 41300
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9534,
"step": 41350
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.967,
"step": 41400
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9712,
"step": 41450
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.966,
"step": 41500
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9675,
"step": 41550
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9728,
"step": 41600
},
{
"epoch": 4.03,
"learning_rate": 3e-06,
"loss": 0.9588,
"step": 41650
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9715,
"step": 41700
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9731,
"step": 41750
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9717,
"step": 41800
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9789,
"step": 41850
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9653,
"step": 41900
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9616,
"step": 41950
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9639,
"step": 42000
},
{
"epoch": 4.04,
"eval_loss": 0.5681419968605042,
"eval_runtime": 39.2115,
"eval_samples_per_second": 4.106,
"eval_steps_per_second": 0.077,
"eval_wer": 29.506699147381244,
"step": 42000
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.956,
"step": 42050
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9585,
"step": 42100
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9598,
"step": 42150
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9657,
"step": 42200
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.975,
"step": 42250
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9592,
"step": 42300
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9659,
"step": 42350
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9548,
"step": 42400
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9556,
"step": 42450
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9593,
"step": 42500
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9671,
"step": 42550
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9665,
"step": 42600
},
{
"epoch": 4.04,
"learning_rate": 3e-06,
"loss": 0.9625,
"step": 42650
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.962,
"step": 42700
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9502,
"step": 42750
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9677,
"step": 42800
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9566,
"step": 42850
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9614,
"step": 42900
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9663,
"step": 42950
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9569,
"step": 43000
},
{
"epoch": 4.05,
"eval_loss": 0.5678820013999939,
"eval_runtime": 39.7895,
"eval_samples_per_second": 4.046,
"eval_steps_per_second": 0.075,
"eval_wer": 29.628501827040193,
"step": 43000
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9676,
"step": 43050
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.966,
"step": 43100
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9577,
"step": 43150
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9658,
"step": 43200
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9663,
"step": 43250
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9638,
"step": 43300
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9635,
"step": 43350
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9631,
"step": 43400
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9602,
"step": 43450
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9619,
"step": 43500
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9521,
"step": 43550
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.96,
"step": 43600
},
{
"epoch": 4.05,
"learning_rate": 3e-06,
"loss": 0.9672,
"step": 43650
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9528,
"step": 43700
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9627,
"step": 43750
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9595,
"step": 43800
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9695,
"step": 43850
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9641,
"step": 43900
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.958,
"step": 43950
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9682,
"step": 44000
},
{
"epoch": 4.06,
"eval_loss": 0.5643439292907715,
"eval_runtime": 39.2385,
"eval_samples_per_second": 4.103,
"eval_steps_per_second": 0.076,
"eval_wer": 29.567600487210722,
"step": 44000
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9527,
"step": 44050
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.96,
"step": 44100
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9641,
"step": 44150
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9748,
"step": 44200
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9659,
"step": 44250
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9565,
"step": 44300
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9606,
"step": 44350
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9561,
"step": 44400
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9578,
"step": 44450
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9665,
"step": 44500
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9657,
"step": 44550
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9535,
"step": 44600
},
{
"epoch": 4.06,
"learning_rate": 3e-06,
"loss": 0.9457,
"step": 44650
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9564,
"step": 44700
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9508,
"step": 44750
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9576,
"step": 44800
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9515,
"step": 44850
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9616,
"step": 44900
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9585,
"step": 44950
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9539,
"step": 45000
},
{
"epoch": 4.07,
"eval_loss": 0.5601379871368408,
"eval_runtime": 39.9689,
"eval_samples_per_second": 4.028,
"eval_steps_per_second": 0.075,
"eval_wer": 29.567600487210722,
"step": 45000
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9527,
"step": 45050
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9553,
"step": 45100
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.962,
"step": 45150
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.96,
"step": 45200
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9679,
"step": 45250
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9522,
"step": 45300
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9536,
"step": 45350
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9557,
"step": 45400
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9501,
"step": 45450
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9559,
"step": 45500
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9539,
"step": 45550
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9556,
"step": 45600
},
{
"epoch": 4.07,
"learning_rate": 3e-06,
"loss": 0.9492,
"step": 45650
},
{
"epoch": 4.08,
"learning_rate": 3e-06,
"loss": 0.9678,
"step": 45700
},
{
"epoch": 4.08,
"learning_rate": 3e-06,
"loss": 0.9634,
"step": 45750
},
{
"epoch": 4.08,
"learning_rate": 3e-06,
"loss": 0.9571,
"step": 45800
},
{
"epoch": 4.08,
"learning_rate": 3e-06,
"loss": 0.9508,
"step": 45850
},
{
"epoch": 4.08,
"learning_rate": 3e-06,
"loss": 0.9468,
"step": 45900
},
{
"epoch": 4.08,
"learning_rate": 3e-06,
"loss": 0.9511,
"step": 45950
},
{
"epoch": 4.08,
"learning_rate": 3e-06,
"loss": 0.946,
"step": 46000
},
{
"epoch": 4.08,
"eval_loss": 0.5562007427215576,
"eval_runtime": 40.2757,
"eval_samples_per_second": 3.997,
"eval_steps_per_second": 0.074,
"eval_wer": 29.71985383678441,
"step": 46000
},
{
"epoch": 4.08,
"learning_rate": 3e-06,
"loss": 0.954,
"step": 46050
},
{
"epoch": 4.08,
"learning_rate": 3e-06,
"loss": 0.9573,
"step": 46100
},
{
"epoch": 4.08,
"learning_rate": 3e-06,
"loss": 0.9508,
"step": 46150
},
{
"epoch": 4.08,
"learning_rate": 3e-06,
"loss": 0.9633,
"step": 46200
},
{
"epoch": 4.08,
"learning_rate": 3e-06,
"loss": 0.9547,
"step": 46250
},
{
"epoch": 4.08,
"learning_rate": 3e-06,
"loss": 0.9515,
"step": 46300
},
{
"epoch": 4.08,
"learning_rate": 3e-06,
"loss": 0.9551,
"step": 46350
},
{
"epoch": 4.08,
"learning_rate": 3e-06,
"loss": 0.9544,
"step": 46400
},
{
"epoch": 5.0,
"learning_rate": 3e-06,
"loss": 0.9554,
"step": 46450
},
{
"epoch": 5.0,
"learning_rate": 3e-06,
"loss": 0.9445,
"step": 46500
},
{
"epoch": 5.0,
"learning_rate": 3e-06,
"loss": 0.9536,
"step": 46550
},
{
"epoch": 5.0,
"learning_rate": 3e-06,
"loss": 0.9375,
"step": 46600
},
{
"epoch": 5.0,
"learning_rate": 3e-06,
"loss": 0.9414,
"step": 46650
},
{
"epoch": 5.0,
"learning_rate": 3e-06,
"loss": 0.9306,
"step": 46700
},
{
"epoch": 5.0,
"learning_rate": 3e-06,
"loss": 0.938,
"step": 46750
},
{
"epoch": 5.0,
"learning_rate": 3e-06,
"loss": 0.9453,
"step": 46800
},
{
"epoch": 5.0,
"learning_rate": 3e-06,
"loss": 0.9366,
"step": 46850
},
{
"epoch": 5.0,
"learning_rate": 3e-06,
"loss": 0.935,
"step": 46900
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9403,
"step": 46950
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9429,
"step": 47000
},
{
"epoch": 5.01,
"eval_loss": 0.5591687560081482,
"eval_runtime": 38.1863,
"eval_samples_per_second": 4.216,
"eval_steps_per_second": 0.079,
"eval_wer": 29.293544457978076,
"step": 47000
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9423,
"step": 47050
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9492,
"step": 47100
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9528,
"step": 47150
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9372,
"step": 47200
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9502,
"step": 47250
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9487,
"step": 47300
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.942,
"step": 47350
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9428,
"step": 47400
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9332,
"step": 47450
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.931,
"step": 47500
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.939,
"step": 47550
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9487,
"step": 47600
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9494,
"step": 47650
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9377,
"step": 47700
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9346,
"step": 47750
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9398,
"step": 47800
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.9444,
"step": 47850
},
{
"epoch": 5.01,
"learning_rate": 3e-06,
"loss": 0.938,
"step": 47900
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.9552,
"step": 47950
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.9462,
"step": 48000
},
{
"epoch": 5.02,
"eval_loss": 0.553955078125,
"eval_runtime": 38.7517,
"eval_samples_per_second": 4.155,
"eval_steps_per_second": 0.077,
"eval_wer": 29.08038976857491,
"step": 48000
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.9412,
"step": 48050
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.9428,
"step": 48100
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.938,
"step": 48150
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.944,
"step": 48200
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.9342,
"step": 48250
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.9285,
"step": 48300
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.9444,
"step": 48350
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.9451,
"step": 48400
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.947,
"step": 48450
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.9345,
"step": 48500
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.9345,
"step": 48550
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.9345,
"step": 48600
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.9357,
"step": 48650
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.9329,
"step": 48700
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.939,
"step": 48750
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.928,
"step": 48800
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.9421,
"step": 48850
},
{
"epoch": 5.02,
"learning_rate": 3e-06,
"loss": 0.9341,
"step": 48900
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9404,
"step": 48950
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9312,
"step": 49000
},
{
"epoch": 5.03,
"eval_loss": 0.5535339713096619,
"eval_runtime": 37.3366,
"eval_samples_per_second": 4.312,
"eval_steps_per_second": 0.08,
"eval_wer": 29.293544457978076,
"step": 49000
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9387,
"step": 49050
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9342,
"step": 49100
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9565,
"step": 49150
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9475,
"step": 49200
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9436,
"step": 49250
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9334,
"step": 49300
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9391,
"step": 49350
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9501,
"step": 49400
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9388,
"step": 49450
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9229,
"step": 49500
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9357,
"step": 49550
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9329,
"step": 49600
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.941,
"step": 49650
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9445,
"step": 49700
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9346,
"step": 49750
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9375,
"step": 49800
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9466,
"step": 49850
},
{
"epoch": 5.03,
"learning_rate": 3e-06,
"loss": 0.9282,
"step": 49900
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9481,
"step": 49950
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9462,
"step": 50000
},
{
"epoch": 5.04,
"eval_loss": 0.5536479949951172,
"eval_runtime": 40.3035,
"eval_samples_per_second": 3.995,
"eval_steps_per_second": 0.074,
"eval_wer": 28.68453105968331,
"step": 50000
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9388,
"step": 50050
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9368,
"step": 50100
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9338,
"step": 50150
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9355,
"step": 50200
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9375,
"step": 50250
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9436,
"step": 50300
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9347,
"step": 50350
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9285,
"step": 50400
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9388,
"step": 50450
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9442,
"step": 50500
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9356,
"step": 50550
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9309,
"step": 50600
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9289,
"step": 50650
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9342,
"step": 50700
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9484,
"step": 50750
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9295,
"step": 50800
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9277,
"step": 50850
},
{
"epoch": 5.04,
"learning_rate": 3e-06,
"loss": 0.9389,
"step": 50900
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9365,
"step": 50950
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.922,
"step": 51000
},
{
"epoch": 5.05,
"eval_loss": 0.5538690686225891,
"eval_runtime": 37.226,
"eval_samples_per_second": 4.325,
"eval_steps_per_second": 0.081,
"eval_wer": 28.714981729598048,
"step": 51000
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9281,
"step": 51050
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9453,
"step": 51100
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9445,
"step": 51150
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9396,
"step": 51200
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9404,
"step": 51250
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9301,
"step": 51300
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9286,
"step": 51350
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9303,
"step": 51400
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9348,
"step": 51450
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9356,
"step": 51500
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9283,
"step": 51550
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9432,
"step": 51600
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9416,
"step": 51650
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9319,
"step": 51700
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.925,
"step": 51750
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.95,
"step": 51800
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9148,
"step": 51850
},
{
"epoch": 5.05,
"learning_rate": 3e-06,
"loss": 0.9348,
"step": 51900
},
{
"epoch": 5.06,
"learning_rate": 3e-06,
"loss": 0.9395,
"step": 51950
},
{
"epoch": 5.06,
"learning_rate": 3e-06,
"loss": 0.9253,
"step": 52000
},
{
"epoch": 5.06,
"eval_loss": 0.5509808659553528,
"eval_runtime": 35.2782,
"eval_samples_per_second": 4.564,
"eval_steps_per_second": 0.085,
"eval_wer": 28.836784409257003,
"step": 52000
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.9213,
"step": 52050
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.9149,
"step": 52100
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.9275,
"step": 52150
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.9333,
"step": 52200
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.9209,
"step": 52250
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.9104,
"step": 52300
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.9305,
"step": 52350
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.9208,
"step": 52400
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.9311,
"step": 52450
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9315,
"step": 52500
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9281,
"step": 52550
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9199,
"step": 52600
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9235,
"step": 52650
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.934,
"step": 52700
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.916,
"step": 52750
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9148,
"step": 52800
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9259,
"step": 52850
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9238,
"step": 52900
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9224,
"step": 52950
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9065,
"step": 53000
},
{
"epoch": 0.01,
"eval_loss": 0.5493320226669312,
"eval_runtime": 40.3992,
"eval_samples_per_second": 3.985,
"eval_steps_per_second": 0.074,
"eval_wer": 28.5931790499391,
"step": 53000
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9211,
"step": 53050
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9196,
"step": 53100
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9215,
"step": 53150
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9177,
"step": 53200
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9131,
"step": 53250
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9228,
"step": 53300
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9292,
"step": 53350
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9221,
"step": 53400
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9316,
"step": 53450
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.9125,
"step": 53500
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9298,
"step": 53550
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9227,
"step": 53600
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9116,
"step": 53650
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9297,
"step": 53700
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.933,
"step": 53750
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9221,
"step": 53800
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.919,
"step": 53850
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9263,
"step": 53900
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9159,
"step": 53950
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9096,
"step": 54000
},
{
"epoch": 0.02,
"eval_loss": 0.5489500164985657,
"eval_runtime": 39.6766,
"eval_samples_per_second": 4.058,
"eval_steps_per_second": 0.076,
"eval_wer": 28.501827040194883,
"step": 54000
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9126,
"step": 54050
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9237,
"step": 54100
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9134,
"step": 54150
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9093,
"step": 54200
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9241,
"step": 54250
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9302,
"step": 54300
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9209,
"step": 54350
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9197,
"step": 54400
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.9119,
"step": 54450
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9085,
"step": 54500
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9109,
"step": 54550
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9121,
"step": 54600
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.927,
"step": 54650
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9125,
"step": 54700
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9125,
"step": 54750
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.917,
"step": 54800
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9195,
"step": 54850
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9271,
"step": 54900
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9256,
"step": 54950
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9329,
"step": 55000
},
{
"epoch": 0.03,
"eval_loss": 0.5482864379882812,
"eval_runtime": 37.1746,
"eval_samples_per_second": 4.331,
"eval_steps_per_second": 0.081,
"eval_wer": 28.288672350791717,
"step": 55000
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9241,
"step": 55050
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9263,
"step": 55100
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9189,
"step": 55150
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9318,
"step": 55200
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9092,
"step": 55250
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9066,
"step": 55300
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9141,
"step": 55350
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9221,
"step": 55400
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9318,
"step": 55450
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9198,
"step": 55500
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9211,
"step": 55550
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9102,
"step": 55600
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9127,
"step": 55650
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9206,
"step": 55700
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9129,
"step": 55750
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.913,
"step": 55800
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9265,
"step": 55850
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9235,
"step": 55900
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9131,
"step": 55950
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9181,
"step": 56000
},
{
"epoch": 0.04,
"eval_loss": 0.5471073985099792,
"eval_runtime": 39.0291,
"eval_samples_per_second": 4.125,
"eval_steps_per_second": 0.077,
"eval_wer": 27.98416565164434,
"step": 56000
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.932,
"step": 56050
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9324,
"step": 56100
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9159,
"step": 56150
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9267,
"step": 56200
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9172,
"step": 56250
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9271,
"step": 56300
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.918,
"step": 56350
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9301,
"step": 56400
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.931,
"step": 56450
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.9095,
"step": 56500
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9182,
"step": 56550
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9115,
"step": 56600
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9187,
"step": 56650
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9336,
"step": 56700
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9269,
"step": 56750
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.915,
"step": 56800
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9263,
"step": 56850
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9049,
"step": 56900
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9169,
"step": 56950
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.914,
"step": 57000
},
{
"epoch": 0.05,
"eval_loss": 0.545651912689209,
"eval_runtime": 38.4894,
"eval_samples_per_second": 4.183,
"eval_steps_per_second": 0.078,
"eval_wer": 28.410475030450673,
"step": 57000
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.92,
"step": 57050
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9217,
"step": 57100
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9247,
"step": 57150
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9188,
"step": 57200
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9181,
"step": 57250
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9123,
"step": 57300
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9202,
"step": 57350
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9264,
"step": 57400
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.9157,
"step": 57450
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9074,
"step": 57500
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9108,
"step": 57550
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.908,
"step": 57600
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9189,
"step": 57650
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9189,
"step": 57700
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9135,
"step": 57750
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9157,
"step": 57800
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9218,
"step": 57850
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9147,
"step": 57900
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9228,
"step": 57950
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9149,
"step": 58000
},
{
"epoch": 0.06,
"eval_loss": 0.5448886752128601,
"eval_runtime": 39.6395,
"eval_samples_per_second": 4.062,
"eval_steps_per_second": 0.076,
"eval_wer": 27.588306942752737,
"step": 58000
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9191,
"step": 58050
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9215,
"step": 58100
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9133,
"step": 58150
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9152,
"step": 58200
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9034,
"step": 58250
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9183,
"step": 58300
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9137,
"step": 58350
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9186,
"step": 58400
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.9166,
"step": 58450
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9153,
"step": 58500
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9159,
"step": 58550
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.921,
"step": 58600
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9109,
"step": 58650
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9079,
"step": 58700
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9129,
"step": 58750
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9141,
"step": 58800
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9191,
"step": 58850
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9078,
"step": 58900
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9067,
"step": 58950
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9092,
"step": 59000
},
{
"epoch": 0.07,
"eval_loss": 0.5405263304710388,
"eval_runtime": 39.5678,
"eval_samples_per_second": 4.069,
"eval_steps_per_second": 0.076,
"eval_wer": 27.831912302070645,
"step": 59000
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9125,
"step": 59050
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9044,
"step": 59100
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9086,
"step": 59150
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9177,
"step": 59200
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9152,
"step": 59250
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9193,
"step": 59300
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9148,
"step": 59350
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9,
"step": 59400
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9084,
"step": 59450
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.9054,
"step": 59500
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8991,
"step": 59550
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.9028,
"step": 59600
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.9068,
"step": 59650
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.9252,
"step": 59700
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.9156,
"step": 59750
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.9235,
"step": 59800
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.903,
"step": 59850
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.9125,
"step": 59900
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.9124,
"step": 59950
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.9101,
"step": 60000
},
{
"epoch": 0.08,
"eval_loss": 0.540199339389801,
"eval_runtime": 38.3011,
"eval_samples_per_second": 4.204,
"eval_steps_per_second": 0.078,
"eval_wer": 27.344701583434833,
"step": 60000
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.9211,
"step": 60050
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.9131,
"step": 60100
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.9119,
"step": 60150
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.9098,
"step": 60200
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.9096,
"step": 60250
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.9053,
"step": 60300
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.9087,
"step": 60350
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.8995,
"step": 60400
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.8805,
"step": 60450
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.8959,
"step": 60500
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.9045,
"step": 60550
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.9092,
"step": 60600
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.9041,
"step": 60650
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.912,
"step": 60700
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.9056,
"step": 60750
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8947,
"step": 60800
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.9007,
"step": 60850
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8982,
"step": 60900
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.9015,
"step": 60950
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.9046,
"step": 61000
},
{
"epoch": 1.01,
"eval_loss": 0.5374026298522949,
"eval_runtime": 40.3306,
"eval_samples_per_second": 3.992,
"eval_steps_per_second": 0.074,
"eval_wer": 27.557856272838006,
"step": 61000
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8987,
"step": 61050
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.9055,
"step": 61100
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.9051,
"step": 61150
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8914,
"step": 61200
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8905,
"step": 61250
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8891,
"step": 61300
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.9029,
"step": 61350
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.9086,
"step": 61400
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.9094,
"step": 61450
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.9161,
"step": 61500
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.9026,
"step": 61550
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8919,
"step": 61600
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8966,
"step": 61650
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8983,
"step": 61700
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8964,
"step": 61750
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8927,
"step": 61800
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.9035,
"step": 61850
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8993,
"step": 61900
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8943,
"step": 61950
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8917,
"step": 62000
},
{
"epoch": 1.02,
"eval_loss": 0.5390424132347107,
"eval_runtime": 37.0496,
"eval_samples_per_second": 4.346,
"eval_steps_per_second": 0.081,
"eval_wer": 27.740560292326432,
"step": 62000
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.9055,
"step": 62050
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.9059,
"step": 62100
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.9021,
"step": 62150
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8927,
"step": 62200
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.901,
"step": 62250
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8946,
"step": 62300
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.9117,
"step": 62350
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8972,
"step": 62400
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8969,
"step": 62450
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8873,
"step": 62500
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.9085,
"step": 62550
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.9078,
"step": 62600
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8987,
"step": 62650
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8939,
"step": 62700
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8953,
"step": 62750
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8906,
"step": 62800
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.9033,
"step": 62850
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8981,
"step": 62900
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.9033,
"step": 62950
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8993,
"step": 63000
},
{
"epoch": 1.03,
"eval_loss": 0.5385776162147522,
"eval_runtime": 39.1791,
"eval_samples_per_second": 4.109,
"eval_steps_per_second": 0.077,
"eval_wer": 27.40560292326431,
"step": 63000
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.9011,
"step": 63050
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.9103,
"step": 63100
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.9001,
"step": 63150
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8957,
"step": 63200
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.9021,
"step": 63250
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8933,
"step": 63300
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8868,
"step": 63350
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.887,
"step": 63400
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8961,
"step": 63450
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.9008,
"step": 63500
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8922,
"step": 63550
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8862,
"step": 63600
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8865,
"step": 63650
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.9023,
"step": 63700
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8946,
"step": 63750
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8954,
"step": 63800
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8864,
"step": 63850
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8879,
"step": 63900
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8884,
"step": 63950
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8875,
"step": 64000
},
{
"epoch": 1.04,
"eval_loss": 0.5361135601997375,
"eval_runtime": 37.6948,
"eval_samples_per_second": 4.271,
"eval_steps_per_second": 0.08,
"eval_wer": 26.857490864799026,
"step": 64000
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.9006,
"step": 64050
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8948,
"step": 64100
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8909,
"step": 64150
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8905,
"step": 64200
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8878,
"step": 64250
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8985,
"step": 64300
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.9024,
"step": 64350
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8999,
"step": 64400
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.9004,
"step": 64450
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8974,
"step": 64500
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8914,
"step": 64550
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.9016,
"step": 64600
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8866,
"step": 64650
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8997,
"step": 64700
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8828,
"step": 64750
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.9024,
"step": 64800
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8879,
"step": 64850
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.907,
"step": 64900
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.9021,
"step": 64950
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8892,
"step": 65000
},
{
"epoch": 1.05,
"eval_loss": 0.5358411073684692,
"eval_runtime": 38.731,
"eval_samples_per_second": 4.157,
"eval_steps_per_second": 0.077,
"eval_wer": 27.344701583434833,
"step": 65000
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.9063,
"step": 65050
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8907,
"step": 65100
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8849,
"step": 65150
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8857,
"step": 65200
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8937,
"step": 65250
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8851,
"step": 65300
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8878,
"step": 65350
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.9036,
"step": 65400
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.91,
"step": 65450
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8864,
"step": 65500
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8888,
"step": 65550
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8933,
"step": 65600
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8898,
"step": 65650
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.9006,
"step": 65700
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8957,
"step": 65750
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8887,
"step": 65800
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8967,
"step": 65850
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.895,
"step": 65900
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8955,
"step": 65950
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8929,
"step": 66000
},
{
"epoch": 1.06,
"eval_loss": 0.5346468091011047,
"eval_runtime": 38.4533,
"eval_samples_per_second": 4.187,
"eval_steps_per_second": 0.078,
"eval_wer": 26.73568818514007,
"step": 66000
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.8919,
"step": 66050
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.8799,
"step": 66100
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.8854,
"step": 66150
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.884,
"step": 66200
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.8742,
"step": 66250
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.8693,
"step": 66300
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.895,
"step": 66350
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.886,
"step": 66400
},
{
"epoch": 0.0,
"learning_rate": 3e-06,
"loss": 0.8899,
"step": 66450
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8952,
"step": 66500
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8939,
"step": 66550
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8787,
"step": 66600
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.881,
"step": 66650
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8903,
"step": 66700
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8749,
"step": 66750
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8794,
"step": 66800
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.899,
"step": 66850
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8862,
"step": 66900
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8834,
"step": 66950
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8703,
"step": 67000
},
{
"epoch": 0.01,
"eval_loss": 0.5332406759262085,
"eval_runtime": 38.9701,
"eval_samples_per_second": 4.131,
"eval_steps_per_second": 0.077,
"eval_wer": 26.82704019488429,
"step": 67000
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.891,
"step": 67050
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8921,
"step": 67100
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8917,
"step": 67150
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8824,
"step": 67200
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8735,
"step": 67250
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8824,
"step": 67300
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8919,
"step": 67350
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8903,
"step": 67400
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.8917,
"step": 67450
},
{
"epoch": 0.01,
"learning_rate": 3e-06,
"loss": 0.872,
"step": 67500
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.89,
"step": 67550
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8824,
"step": 67600
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8718,
"step": 67650
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8881,
"step": 67700
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8933,
"step": 67750
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8834,
"step": 67800
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.878,
"step": 67850
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8871,
"step": 67900
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8799,
"step": 67950
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8709,
"step": 68000
},
{
"epoch": 0.02,
"eval_loss": 0.5335850715637207,
"eval_runtime": 40.527,
"eval_samples_per_second": 3.973,
"eval_steps_per_second": 0.074,
"eval_wer": 26.705237515225335,
"step": 68000
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8754,
"step": 68050
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8951,
"step": 68100
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8765,
"step": 68150
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8698,
"step": 68200
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8834,
"step": 68250
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.891,
"step": 68300
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8845,
"step": 68350
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8918,
"step": 68400
},
{
"epoch": 0.02,
"learning_rate": 3e-06,
"loss": 0.8838,
"step": 68450
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8798,
"step": 68500
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8744,
"step": 68550
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.878,
"step": 68600
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8953,
"step": 68650
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8737,
"step": 68700
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8733,
"step": 68750
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8783,
"step": 68800
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8885,
"step": 68850
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8957,
"step": 68900
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8853,
"step": 68950
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8917,
"step": 69000
},
{
"epoch": 0.03,
"eval_loss": 0.5328567028045654,
"eval_runtime": 39.7121,
"eval_samples_per_second": 4.054,
"eval_steps_per_second": 0.076,
"eval_wer": 27.070645554202194,
"step": 69000
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8826,
"step": 69050
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8879,
"step": 69100
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8796,
"step": 69150
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8915,
"step": 69200
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8707,
"step": 69250
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8722,
"step": 69300
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8864,
"step": 69350
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.8908,
"step": 69400
},
{
"epoch": 0.03,
"learning_rate": 3e-06,
"loss": 0.9004,
"step": 69450
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.89,
"step": 69500
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.885,
"step": 69550
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.8716,
"step": 69600
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.876,
"step": 69650
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.8831,
"step": 69700
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.8746,
"step": 69750
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.8718,
"step": 69800
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.8868,
"step": 69850
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.8855,
"step": 69900
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.884,
"step": 69950
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.8867,
"step": 70000
},
{
"epoch": 0.04,
"eval_loss": 0.5323489904403687,
"eval_runtime": 38.5858,
"eval_samples_per_second": 4.173,
"eval_steps_per_second": 0.078,
"eval_wer": 26.33982947624848,
"step": 70000
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.892,
"step": 70050
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.8939,
"step": 70100
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.8801,
"step": 70150
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.8955,
"step": 70200
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.8804,
"step": 70250
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.8877,
"step": 70300
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.887,
"step": 70350
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.8989,
"step": 70400
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.8997,
"step": 70450
},
{
"epoch": 0.04,
"learning_rate": 3e-06,
"loss": 0.873,
"step": 70500
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8836,
"step": 70550
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8801,
"step": 70600
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8835,
"step": 70650
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8997,
"step": 70700
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8984,
"step": 70750
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8789,
"step": 70800
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8869,
"step": 70850
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8683,
"step": 70900
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8801,
"step": 70950
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8778,
"step": 71000
},
{
"epoch": 0.05,
"eval_loss": 0.5315227508544922,
"eval_runtime": 40.5435,
"eval_samples_per_second": 3.971,
"eval_steps_per_second": 0.074,
"eval_wer": 27.28380024360536,
"step": 71000
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8828,
"step": 71050
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8836,
"step": 71100
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8853,
"step": 71150
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8885,
"step": 71200
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8895,
"step": 71250
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.884,
"step": 71300
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8843,
"step": 71350
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8881,
"step": 71400
},
{
"epoch": 0.05,
"learning_rate": 3e-06,
"loss": 0.8797,
"step": 71450
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8716,
"step": 71500
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8747,
"step": 71550
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8715,
"step": 71600
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8816,
"step": 71650
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8806,
"step": 71700
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8751,
"step": 71750
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8798,
"step": 71800
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8906,
"step": 71850
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8773,
"step": 71900
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8835,
"step": 71950
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8757,
"step": 72000
},
{
"epoch": 0.06,
"eval_loss": 0.5316939353942871,
"eval_runtime": 41.0725,
"eval_samples_per_second": 3.92,
"eval_steps_per_second": 0.073,
"eval_wer": 26.248477466504262,
"step": 72000
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8853,
"step": 72050
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8912,
"step": 72100
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8848,
"step": 72150
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8867,
"step": 72200
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8751,
"step": 72250
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8824,
"step": 72300
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8771,
"step": 72350
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8824,
"step": 72400
},
{
"epoch": 0.06,
"learning_rate": 3e-06,
"loss": 0.8848,
"step": 72450
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8821,
"step": 72500
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8774,
"step": 72550
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8841,
"step": 72600
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8749,
"step": 72650
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8716,
"step": 72700
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8805,
"step": 72750
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8838,
"step": 72800
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8822,
"step": 72850
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.872,
"step": 72900
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8699,
"step": 72950
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8726,
"step": 73000
},
{
"epoch": 0.07,
"eval_loss": 0.5268862843513489,
"eval_runtime": 38.1185,
"eval_samples_per_second": 4.224,
"eval_steps_per_second": 0.079,
"eval_wer": 26.64433617539586,
"step": 73000
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8761,
"step": 73050
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8675,
"step": 73100
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8775,
"step": 73150
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.889,
"step": 73200
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8847,
"step": 73250
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8843,
"step": 73300
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8765,
"step": 73350
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8638,
"step": 73400
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8724,
"step": 73450
},
{
"epoch": 0.07,
"learning_rate": 3e-06,
"loss": 0.8697,
"step": 73500
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8634,
"step": 73550
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8669,
"step": 73600
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8701,
"step": 73650
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8936,
"step": 73700
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8845,
"step": 73750
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8858,
"step": 73800
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8662,
"step": 73850
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8769,
"step": 73900
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8843,
"step": 73950
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8792,
"step": 74000
},
{
"epoch": 0.08,
"eval_loss": 0.526807427406311,
"eval_runtime": 38.075,
"eval_samples_per_second": 4.229,
"eval_steps_per_second": 0.079,
"eval_wer": 26.15712545676005,
"step": 74000
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8865,
"step": 74050
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8765,
"step": 74100
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8757,
"step": 74150
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8743,
"step": 74200
},
{
"epoch": 0.08,
"learning_rate": 3e-06,
"loss": 0.8726,
"step": 74250
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.869,
"step": 74300
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.8727,
"step": 74350
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.8629,
"step": 74400
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.8445,
"step": 74450
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.8617,
"step": 74500
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.8703,
"step": 74550
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.8754,
"step": 74600
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.8683,
"step": 74650
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.8766,
"step": 74700
},
{
"epoch": 1.0,
"learning_rate": 3e-06,
"loss": 0.8712,
"step": 74750
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8589,
"step": 74800
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8652,
"step": 74850
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8646,
"step": 74900
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8675,
"step": 74950
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8706,
"step": 75000
},
{
"epoch": 1.01,
"eval_loss": 0.5247220993041992,
"eval_runtime": 39.3966,
"eval_samples_per_second": 4.087,
"eval_steps_per_second": 0.076,
"eval_wer": 26.15712545676005,
"step": 75000
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8645,
"step": 75050
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8716,
"step": 75100
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8705,
"step": 75150
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.858,
"step": 75200
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8568,
"step": 75250
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8549,
"step": 75300
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8668,
"step": 75350
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8727,
"step": 75400
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8738,
"step": 75450
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.88,
"step": 75500
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8669,
"step": 75550
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8576,
"step": 75600
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8612,
"step": 75650
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8623,
"step": 75700
},
{
"epoch": 1.01,
"learning_rate": 3e-06,
"loss": 0.8613,
"step": 75750
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8575,
"step": 75800
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8687,
"step": 75850
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8654,
"step": 75900
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8602,
"step": 75950
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8585,
"step": 76000
},
{
"epoch": 1.02,
"eval_loss": 0.5264820456504822,
"eval_runtime": 38.9874,
"eval_samples_per_second": 4.13,
"eval_steps_per_second": 0.077,
"eval_wer": 26.370280146163218,
"step": 76000
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8717,
"step": 76050
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8724,
"step": 76100
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8691,
"step": 76150
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8587,
"step": 76200
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8665,
"step": 76250
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8613,
"step": 76300
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8781,
"step": 76350
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8639,
"step": 76400
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8638,
"step": 76450
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8534,
"step": 76500
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8729,
"step": 76550
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8733,
"step": 76600
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.864,
"step": 76650
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8605,
"step": 76700
},
{
"epoch": 1.02,
"learning_rate": 3e-06,
"loss": 0.8624,
"step": 76750
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8575,
"step": 76800
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8696,
"step": 76850
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8636,
"step": 76900
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8684,
"step": 76950
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8659,
"step": 77000
},
{
"epoch": 1.03,
"eval_loss": 0.5261924862861633,
"eval_runtime": 39.7957,
"eval_samples_per_second": 4.046,
"eval_steps_per_second": 0.075,
"eval_wer": 26.73568818514007,
"step": 77000
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8672,
"step": 77050
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8756,
"step": 77100
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8668,
"step": 77150
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8619,
"step": 77200
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8676,
"step": 77250
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8594,
"step": 77300
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8534,
"step": 77350
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8539,
"step": 77400
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8632,
"step": 77450
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8672,
"step": 77500
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8589,
"step": 77550
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8525,
"step": 77600
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8528,
"step": 77650
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.868,
"step": 77700
},
{
"epoch": 1.03,
"learning_rate": 3e-06,
"loss": 0.8608,
"step": 77750
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8622,
"step": 77800
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8536,
"step": 77850
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8548,
"step": 77900
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8552,
"step": 77950
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8551,
"step": 78000
},
{
"epoch": 1.04,
"eval_loss": 0.5248793363571167,
"eval_runtime": 38.0291,
"eval_samples_per_second": 4.234,
"eval_steps_per_second": 0.079,
"eval_wer": 26.065773447015832,
"step": 78000
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.868,
"step": 78050
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8623,
"step": 78100
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8586,
"step": 78150
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8573,
"step": 78200
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8553,
"step": 78250
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8659,
"step": 78300
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8696,
"step": 78350
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8676,
"step": 78400
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8669,
"step": 78450
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8631,
"step": 78500
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8586,
"step": 78550
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8691,
"step": 78600
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8537,
"step": 78650
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8671,
"step": 78700
},
{
"epoch": 1.04,
"learning_rate": 3e-06,
"loss": 0.8505,
"step": 78750
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8684,
"step": 78800
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8555,
"step": 78850
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8747,
"step": 78900
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8695,
"step": 78950
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8572,
"step": 79000
},
{
"epoch": 1.05,
"eval_loss": 0.5249020457267761,
"eval_runtime": 38.8753,
"eval_samples_per_second": 4.141,
"eval_steps_per_second": 0.077,
"eval_wer": 26.278928136419,
"step": 79000
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.874,
"step": 79050
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8587,
"step": 79100
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8527,
"step": 79150
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.853,
"step": 79200
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8614,
"step": 79250
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8535,
"step": 79300
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.856,
"step": 79350
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.871,
"step": 79400
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.877,
"step": 79450
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.854,
"step": 79500
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.856,
"step": 79550
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8606,
"step": 79600
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8567,
"step": 79650
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8673,
"step": 79700
},
{
"epoch": 1.05,
"learning_rate": 3e-06,
"loss": 0.8626,
"step": 79750
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8571,
"step": 79800
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8651,
"step": 79850
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8625,
"step": 79900
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8639,
"step": 79950
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8612,
"step": 80000
},
{
"epoch": 1.06,
"eval_loss": 0.5234566330909729,
"eval_runtime": 39.62,
"eval_samples_per_second": 4.064,
"eval_steps_per_second": 0.076,
"eval_wer": 25.761266747868454,
"step": 80000
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8742,
"step": 80050
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8658,
"step": 80100
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8611,
"step": 80150
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8718,
"step": 80200
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8697,
"step": 80250
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8722,
"step": 80300
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.873,
"step": 80350
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8662,
"step": 80400
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8656,
"step": 80450
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.87,
"step": 80500
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8658,
"step": 80550
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8703,
"step": 80600
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8718,
"step": 80650
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8589,
"step": 80700
},
{
"epoch": 1.06,
"learning_rate": 3e-06,
"loss": 0.8676,
"step": 80750
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8589,
"step": 80800
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8624,
"step": 80850
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8645,
"step": 80900
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8727,
"step": 80950
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8598,
"step": 81000
},
{
"epoch": 1.07,
"eval_loss": 0.5208215117454529,
"eval_runtime": 38.4397,
"eval_samples_per_second": 4.188,
"eval_steps_per_second": 0.078,
"eval_wer": 25.700365408038977,
"step": 81000
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.882,
"step": 81050
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8737,
"step": 81100
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8689,
"step": 81150
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8631,
"step": 81200
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8722,
"step": 81250
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8688,
"step": 81300
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8662,
"step": 81350
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8627,
"step": 81400
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8671,
"step": 81450
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8573,
"step": 81500
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.861,
"step": 81550
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.867,
"step": 81600
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8639,
"step": 81650
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8607,
"step": 81700
},
{
"epoch": 1.07,
"learning_rate": 3e-06,
"loss": 0.8602,
"step": 81750
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.8719,
"step": 81800
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.8637,
"step": 81850
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.8598,
"step": 81900
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.867,
"step": 81950
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.8686,
"step": 82000
},
{
"epoch": 1.08,
"eval_loss": 0.5214089155197144,
"eval_runtime": 37.0595,
"eval_samples_per_second": 4.344,
"eval_steps_per_second": 0.081,
"eval_wer": 25.700365408038977,
"step": 82000
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.8712,
"step": 82050
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.8656,
"step": 82100
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.8677,
"step": 82150
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.8656,
"step": 82200
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.8651,
"step": 82250
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.874,
"step": 82300
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.8735,
"step": 82350
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.8637,
"step": 82400
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.8657,
"step": 82450
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.8685,
"step": 82500
},
{
"epoch": 1.08,
"learning_rate": 3e-06,
"loss": 0.8534,
"step": 82550
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 0.8645,
"step": 82600
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 0.8448,
"step": 82650
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 0.8565,
"step": 82700
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 0.8537,
"step": 82750
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 0.8496,
"step": 82800
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 0.8512,
"step": 82850
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 0.8483,
"step": 82900
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 0.8476,
"step": 82950
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 0.8503,
"step": 83000
},
{
"epoch": 2.0,
"eval_loss": 0.5213810205459595,
"eval_runtime": 38.3305,
"eval_samples_per_second": 4.2,
"eval_steps_per_second": 0.078,
"eval_wer": 25.700365408038977,
"step": 83000
},
{
"epoch": 2.0,
"learning_rate": 3e-06,
"loss": 0.85,
"step": 83050
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.848,
"step": 83100
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8498,
"step": 83150
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8523,
"step": 83200
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8537,
"step": 83250
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8542,
"step": 83300
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8519,
"step": 83350
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.857,
"step": 83400
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8636,
"step": 83450
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8408,
"step": 83500
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8688,
"step": 83550
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8539,
"step": 83600
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8406,
"step": 83650
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8532,
"step": 83700
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.86,
"step": 83750
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8591,
"step": 83800
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8469,
"step": 83850
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8537,
"step": 83900
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.848,
"step": 83950
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8545,
"step": 84000
},
{
"epoch": 2.01,
"eval_loss": 0.5214766263961792,
"eval_runtime": 53.964,
"eval_samples_per_second": 2.983,
"eval_steps_per_second": 0.056,
"eval_wer": 28.227771010962243,
"step": 84000
},
{
"epoch": 2.01,
"learning_rate": 3e-06,
"loss": 0.8418,
"step": 84050
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8581,
"step": 84100
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.851,
"step": 84150
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8384,
"step": 84200
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8519,
"step": 84250
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.859,
"step": 84300
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8476,
"step": 84350
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8543,
"step": 84400
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8462,
"step": 84450
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8383,
"step": 84500
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8431,
"step": 84550
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.852,
"step": 84600
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8607,
"step": 84650
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8469,
"step": 84700
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8391,
"step": 84750
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8567,
"step": 84800
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8409,
"step": 84850
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8509,
"step": 84900
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8606,
"step": 84950
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8594,
"step": 85000
},
{
"epoch": 2.02,
"eval_loss": 0.5186431407928467,
"eval_runtime": 39.735,
"eval_samples_per_second": 4.052,
"eval_steps_per_second": 0.076,
"eval_wer": 25.669914738124238,
"step": 85000
},
{
"epoch": 2.02,
"learning_rate": 3e-06,
"loss": 0.8663,
"step": 85050
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8503,
"step": 85100
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.848,
"step": 85150
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8507,
"step": 85200
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8559,
"step": 85250
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8536,
"step": 85300
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8562,
"step": 85350
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.843,
"step": 85400
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8628,
"step": 85450
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8418,
"step": 85500
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8493,
"step": 85550
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8544,
"step": 85600
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8529,
"step": 85650
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8573,
"step": 85700
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8541,
"step": 85750
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8428,
"step": 85800
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8432,
"step": 85850
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8402,
"step": 85900
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8465,
"step": 85950
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.86,
"step": 86000
},
{
"epoch": 2.03,
"eval_loss": 0.5196462273597717,
"eval_runtime": 40.0681,
"eval_samples_per_second": 4.018,
"eval_steps_per_second": 0.075,
"eval_wer": 25.57856272838002,
"step": 86000
},
{
"epoch": 2.03,
"learning_rate": 3e-06,
"loss": 0.8511,
"step": 86050
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.845,
"step": 86100
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8432,
"step": 86150
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8493,
"step": 86200
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8467,
"step": 86250
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8396,
"step": 86300
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8501,
"step": 86350
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8504,
"step": 86400
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8699,
"step": 86450
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8479,
"step": 86500
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8433,
"step": 86550
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8577,
"step": 86600
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8428,
"step": 86650
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8472,
"step": 86700
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8506,
"step": 86750
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8485,
"step": 86800
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8391,
"step": 86850
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8391,
"step": 86900
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8627,
"step": 86950
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8514,
"step": 87000
},
{
"epoch": 2.04,
"eval_loss": 0.5203261971473694,
"eval_runtime": 39.1863,
"eval_samples_per_second": 4.109,
"eval_steps_per_second": 0.077,
"eval_wer": 25.182704019488426,
"step": 87000
},
{
"epoch": 2.04,
"learning_rate": 3e-06,
"loss": 0.8551,
"step": 87050
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8441,
"step": 87100
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8528,
"step": 87150
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8619,
"step": 87200
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8434,
"step": 87250
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.846,
"step": 87300
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8567,
"step": 87350
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8502,
"step": 87400
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8531,
"step": 87450
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8515,
"step": 87500
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8344,
"step": 87550
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8511,
"step": 87600
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8474,
"step": 87650
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8474,
"step": 87700
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8569,
"step": 87750
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8505,
"step": 87800
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8571,
"step": 87850
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8469,
"step": 87900
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8564,
"step": 87950
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8505,
"step": 88000
},
{
"epoch": 2.05,
"eval_loss": 0.5163885354995728,
"eval_runtime": 50.9215,
"eval_samples_per_second": 3.162,
"eval_steps_per_second": 0.059,
"eval_wer": 28.01461632155907,
"step": 88000
},
{
"epoch": 2.05,
"learning_rate": 3e-06,
"loss": 0.8538,
"step": 88050
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8592,
"step": 88100
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.851,
"step": 88150
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8455,
"step": 88200
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8412,
"step": 88250
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8507,
"step": 88300
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8462,
"step": 88350
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8452,
"step": 88400
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8406,
"step": 88450
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8348,
"step": 88500
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8376,
"step": 88550
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8336,
"step": 88600
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.848,
"step": 88650
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8507,
"step": 88700
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8512,
"step": 88750
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8441,
"step": 88800
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8459,
"step": 88850
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8484,
"step": 88900
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8389,
"step": 88950
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8512,
"step": 89000
},
{
"epoch": 2.06,
"eval_loss": 0.5173874497413635,
"eval_runtime": 39.69,
"eval_samples_per_second": 4.056,
"eval_steps_per_second": 0.076,
"eval_wer": 25.091352009744217,
"step": 89000
},
{
"epoch": 2.06,
"learning_rate": 3e-06,
"loss": 0.8477,
"step": 89050
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8642,
"step": 89100
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8497,
"step": 89150
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8489,
"step": 89200
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8383,
"step": 89250
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8441,
"step": 89300
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8523,
"step": 89350
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.836,
"step": 89400
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8397,
"step": 89450
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8432,
"step": 89500
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8383,
"step": 89550
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8475,
"step": 89600
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8469,
"step": 89650
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8537,
"step": 89700
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8516,
"step": 89750
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8452,
"step": 89800
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.847,
"step": 89850
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8373,
"step": 89900
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8446,
"step": 89950
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8495,
"step": 90000
},
{
"epoch": 2.07,
"eval_loss": 0.5141222476959229,
"eval_runtime": 38.4488,
"eval_samples_per_second": 4.187,
"eval_steps_per_second": 0.078,
"eval_wer": 25.54811205846529,
"step": 90000
},
{
"epoch": 2.07,
"learning_rate": 3e-06,
"loss": 0.8508,
"step": 90050
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.8408,
"step": 90100
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.8439,
"step": 90150
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.8476,
"step": 90200
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.8434,
"step": 90250
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.8535,
"step": 90300
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.8414,
"step": 90350
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.853,
"step": 90400
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.8436,
"step": 90450
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.8366,
"step": 90500
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.8387,
"step": 90550
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.8421,
"step": 90600
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.8371,
"step": 90650
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.8542,
"step": 90700
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.8543,
"step": 90750
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.8352,
"step": 90800
},
{
"epoch": 2.08,
"learning_rate": 3e-06,
"loss": 0.8383,
"step": 90850
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 0.841,
"step": 90900
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 0.839,
"step": 90950
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 0.8381,
"step": 91000
},
{
"epoch": 3.0,
"eval_loss": 0.5129591226577759,
"eval_runtime": 38.5934,
"eval_samples_per_second": 4.172,
"eval_steps_per_second": 0.078,
"eval_wer": 24.96954933008526,
"step": 91000
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 0.828,
"step": 91050
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 0.8258,
"step": 91100
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 0.8379,
"step": 91150
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 0.8315,
"step": 91200
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 0.8401,
"step": 91250
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 0.8444,
"step": 91300
},
{
"epoch": 3.0,
"learning_rate": 3e-06,
"loss": 0.843,
"step": 91350
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8325,
"step": 91400
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8416,
"step": 91450
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8468,
"step": 91500
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8376,
"step": 91550
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8431,
"step": 91600
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8307,
"step": 91650
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.83,
"step": 91700
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8404,
"step": 91750
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8429,
"step": 91800
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8328,
"step": 91850
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.83,
"step": 91900
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8256,
"step": 91950
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8253,
"step": 92000
},
{
"epoch": 3.01,
"eval_loss": 0.5146694779396057,
"eval_runtime": 35.1807,
"eval_samples_per_second": 4.576,
"eval_steps_per_second": 0.085,
"eval_wer": 25.57856272838002,
"step": 92000
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8357,
"step": 92050
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8417,
"step": 92100
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8426,
"step": 92150
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8331,
"step": 92200
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8215,
"step": 92250
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8274,
"step": 92300
},
{
"epoch": 3.01,
"learning_rate": 3e-06,
"loss": 0.8268,
"step": 92350
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8322,
"step": 92400
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.84,
"step": 92450
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8407,
"step": 92500
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8404,
"step": 92550
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8395,
"step": 92600
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8346,
"step": 92650
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8292,
"step": 92700
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8389,
"step": 92750
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8261,
"step": 92800
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8261,
"step": 92850
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8343,
"step": 92900
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8294,
"step": 92950
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8387,
"step": 93000
},
{
"epoch": 3.02,
"eval_loss": 0.5168384313583374,
"eval_runtime": 38.8899,
"eval_samples_per_second": 4.14,
"eval_steps_per_second": 0.077,
"eval_wer": 24.908647990255787,
"step": 93000
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8296,
"step": 93050
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8306,
"step": 93100
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8364,
"step": 93150
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8453,
"step": 93200
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8191,
"step": 93250
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8257,
"step": 93300
},
{
"epoch": 3.02,
"learning_rate": 3e-06,
"loss": 0.8376,
"step": 93350
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8242,
"step": 93400
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8238,
"step": 93450
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8319,
"step": 93500
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8355,
"step": 93550
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8372,
"step": 93600
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8326,
"step": 93650
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8339,
"step": 93700
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.838,
"step": 93750
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8365,
"step": 93800
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8359,
"step": 93850
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8385,
"step": 93900
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8357,
"step": 93950
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8425,
"step": 94000
},
{
"epoch": 3.03,
"eval_loss": 0.5134768486022949,
"eval_runtime": 37.6084,
"eval_samples_per_second": 4.281,
"eval_steps_per_second": 0.08,
"eval_wer": 25.243605359317904,
"step": 94000
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8331,
"step": 94050
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8438,
"step": 94100
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8389,
"step": 94150
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8274,
"step": 94200
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8254,
"step": 94250
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8322,
"step": 94300
},
{
"epoch": 3.03,
"learning_rate": 3e-06,
"loss": 0.8471,
"step": 94350
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8358,
"step": 94400
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8337,
"step": 94450
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8381,
"step": 94500
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8384,
"step": 94550
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8286,
"step": 94600
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8332,
"step": 94650
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8362,
"step": 94700
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8306,
"step": 94750
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8367,
"step": 94800
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8349,
"step": 94850
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8286,
"step": 94900
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8249,
"step": 94950
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8339,
"step": 95000
},
{
"epoch": 3.04,
"eval_loss": 0.5161515474319458,
"eval_runtime": 40.2582,
"eval_samples_per_second": 3.999,
"eval_steps_per_second": 0.075,
"eval_wer": 25.669914738124238,
"step": 95000
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8428,
"step": 95050
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8315,
"step": 95100
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8442,
"step": 95150
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8305,
"step": 95200
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8305,
"step": 95250
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8485,
"step": 95300
},
{
"epoch": 3.04,
"learning_rate": 3e-06,
"loss": 0.8374,
"step": 95350
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8432,
"step": 95400
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8406,
"step": 95450
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8399,
"step": 95500
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8377,
"step": 95550
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8297,
"step": 95600
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8499,
"step": 95650
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8362,
"step": 95700
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8404,
"step": 95750
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8365,
"step": 95800
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8375,
"step": 95850
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.832,
"step": 95900
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8343,
"step": 95950
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8402,
"step": 96000
},
{
"epoch": 3.05,
"eval_loss": 0.5146769881248474,
"eval_runtime": 36.7672,
"eval_samples_per_second": 4.379,
"eval_steps_per_second": 0.082,
"eval_wer": 25.730816077953715,
"step": 96000
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8398,
"step": 96050
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8369,
"step": 96100
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8496,
"step": 96150
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8433,
"step": 96200
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8453,
"step": 96250
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8381,
"step": 96300
},
{
"epoch": 3.05,
"learning_rate": 3e-06,
"loss": 0.8324,
"step": 96350
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8411,
"step": 96400
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8311,
"step": 96450
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8378,
"step": 96500
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8364,
"step": 96550
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8361,
"step": 96600
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8368,
"step": 96650
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8367,
"step": 96700
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.833,
"step": 96750
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8383,
"step": 96800
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8352,
"step": 96850
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8421,
"step": 96900
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8359,
"step": 96950
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8396,
"step": 97000
},
{
"epoch": 3.06,
"eval_loss": 0.5142699480056763,
"eval_runtime": 37.3451,
"eval_samples_per_second": 4.311,
"eval_steps_per_second": 0.08,
"eval_wer": 25.669914738124238,
"step": 97000
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8277,
"step": 97050
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8331,
"step": 97100
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.825,
"step": 97150
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8222,
"step": 97200
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8335,
"step": 97250
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8344,
"step": 97300
},
{
"epoch": 3.06,
"learning_rate": 3e-06,
"loss": 0.8327,
"step": 97350
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8256,
"step": 97400
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8365,
"step": 97450
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8245,
"step": 97500
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8234,
"step": 97550
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8386,
"step": 97600
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8338,
"step": 97650
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8403,
"step": 97700
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8465,
"step": 97750
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.836,
"step": 97800
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8236,
"step": 97850
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8387,
"step": 97900
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8309,
"step": 97950
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8432,
"step": 98000
},
{
"epoch": 3.07,
"eval_loss": 0.5099794864654541,
"eval_runtime": 38.2718,
"eval_samples_per_second": 4.207,
"eval_steps_per_second": 0.078,
"eval_wer": 24.878197320341048,
"step": 98000
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8324,
"step": 98050
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8328,
"step": 98100
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8281,
"step": 98150
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8282,
"step": 98200
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8376,
"step": 98250
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8453,
"step": 98300
},
{
"epoch": 3.07,
"learning_rate": 3e-06,
"loss": 0.8347,
"step": 98350
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.828,
"step": 98400
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.8273,
"step": 98450
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.8421,
"step": 98500
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.8237,
"step": 98550
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.8207,
"step": 98600
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.8481,
"step": 98650
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.8414,
"step": 98700
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.8472,
"step": 98750
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.8257,
"step": 98800
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.8283,
"step": 98850
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.8336,
"step": 98900
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.832,
"step": 98950
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.844,
"step": 99000
},
{
"epoch": 3.08,
"eval_loss": 0.5100468397140503,
"eval_runtime": 36.7593,
"eval_samples_per_second": 4.38,
"eval_steps_per_second": 0.082,
"eval_wer": 25.060901339829478,
"step": 99000
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.839,
"step": 99050
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.838,
"step": 99100
},
{
"epoch": 3.08,
"learning_rate": 3e-06,
"loss": 0.8347,
"step": 99150
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.8243,
"step": 99200
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.8305,
"step": 99250
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.8251,
"step": 99300
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.8218,
"step": 99350
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.8176,
"step": 99400
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.8345,
"step": 99450
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.8259,
"step": 99500
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.8298,
"step": 99550
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.8183,
"step": 99600
},
{
"epoch": 4.0,
"learning_rate": 3e-06,
"loss": 0.83,
"step": 99650
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.8267,
"step": 99700
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.8235,
"step": 99750
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.8223,
"step": 99800
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.8331,
"step": 99850
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.82,
"step": 99900
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.8196,
"step": 99950
},
{
"epoch": 4.01,
"learning_rate": 3e-06,
"loss": 0.8333,
"step": 100000
},
{
"epoch": 4.01,
"eval_loss": 0.512780487537384,
"eval_runtime": 38.9453,
"eval_samples_per_second": 4.134,
"eval_steps_per_second": 0.077,
"eval_wer": 24.96954933008526,
"step": 100000
},
{
"epoch": 4.01,
"step": 100000,
"total_flos": 3.151209909694464e+20,
"train_loss": 0.2913371723175049,
"train_runtime": 623384.3788,
"train_samples_per_second": 41.066,
"train_steps_per_second": 0.16
}
],
"max_steps": 100000,
"num_train_epochs": 9223372036854775807,
"total_flos": 3.151209909694464e+20,
"trial_name": null,
"trial_params": null
}