{
  "best_metric": 4.054176686782522,
  "best_model_checkpoint": "./raid/liyan/whisper/whisper-medium-mls_sjt/checkpoint-7500",
  "epoch": 2.5737817433081673,
  "global_step": 7500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.4039,
      "step": 25
    },
    {
      "epoch": 0.02,
      "learning_rate": 9.800000000000001e-07,
      "loss": 0.3669,
      "step": 50
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.46e-06,
      "loss": 0.3318,
      "step": 75
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9600000000000003e-06,
      "loss": 0.3159,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.46e-06,
      "loss": 0.296,
      "step": 125
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.96e-06,
      "loss": 0.3047,
      "step": 150
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.46e-06,
      "loss": 0.2821,
      "step": 175
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.96e-06,
      "loss": 0.2893,
      "step": 200
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.4600000000000005e-06,
      "loss": 0.2679,
      "step": 225
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.960000000000001e-06,
      "loss": 0.2591,
      "step": 250
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.460000000000001e-06,
      "loss": 0.2453,
      "step": 275
    },
    {
      "epoch": 0.1,
      "learning_rate": 5.9600000000000005e-06,
      "loss": 0.2507,
      "step": 300
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.460000000000001e-06,
      "loss": 0.2435,
      "step": 325
    },
    {
      "epoch": 0.12,
      "learning_rate": 6.96e-06,
      "loss": 0.2305,
      "step": 350
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.4600000000000006e-06,
      "loss": 0.2224,
      "step": 375
    },
    {
      "epoch": 0.14,
      "learning_rate": 7.960000000000002e-06,
      "loss": 0.2443,
      "step": 400
    },
    {
      "epoch": 0.15,
      "learning_rate": 8.46e-06,
      "loss": 0.2443,
      "step": 425
    },
    {
      "epoch": 0.15,
      "learning_rate": 8.96e-06,
      "loss": 0.2292,
      "step": 450
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.460000000000001e-06,
      "loss": 0.2445,
      "step": 475
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.960000000000001e-06,
      "loss": 0.2463,
      "step": 500
    },
    {
      "epoch": 0.17,
      "eval_loss": 0.14288775622844696,
      "eval_runtime": 13972.8923,
      "eval_samples_per_second": 1.601,
      "eval_steps_per_second": 0.067,
      "eval_wer": 5.199872952493307,
      "step": 500
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.99220338983051e-06,
      "loss": 0.2158,
      "step": 525
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.983728813559322e-06,
      "loss": 0.2302,
      "step": 550
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.975254237288137e-06,
      "loss": 0.2223,
      "step": 575
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.96677966101695e-06,
      "loss": 0.216,
      "step": 600
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.958305084745764e-06,
      "loss": 0.2111,
      "step": 625
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.949830508474577e-06,
      "loss": 0.2208,
      "step": 650
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.94135593220339e-06,
      "loss": 0.2254,
      "step": 675
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.932881355932205e-06,
      "loss": 0.2008,
      "step": 700
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.924406779661018e-06,
      "loss": 0.2284,
      "step": 725
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.915932203389831e-06,
      "loss": 0.2193,
      "step": 750
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.907457627118646e-06,
      "loss": 0.2049,
      "step": 775
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.898983050847459e-06,
      "loss": 0.2048,
      "step": 800
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.890508474576272e-06,
      "loss": 0.2272,
      "step": 825
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.882033898305086e-06,
      "loss": 0.2096,
      "step": 850
    },
    {
      "epoch": 0.3,
      "learning_rate": 9.8735593220339e-06,
      "loss": 0.2078,
      "step": 875
    },
    {
      "epoch": 0.31,
      "learning_rate": 9.865084745762712e-06,
      "loss": 0.206,
      "step": 900
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.856610169491525e-06,
      "loss": 0.211,
      "step": 925
    },
    {
      "epoch": 0.33,
      "learning_rate": 9.84813559322034e-06,
      "loss": 0.2239,
      "step": 950
    },
    {
      "epoch": 0.33,
      "learning_rate": 9.839661016949153e-06,
      "loss": 0.2099,
      "step": 975
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.831186440677966e-06,
      "loss": 0.208,
      "step": 1000
    },
    {
      "epoch": 0.34,
      "eval_loss": 0.1417330801486969,
      "eval_runtime": 14048.5832,
      "eval_samples_per_second": 1.592,
      "eval_steps_per_second": 0.066,
      "eval_wer": 5.172194745678116,
      "step": 1000
    },
    {
      "epoch": 0.35,
      "learning_rate": 9.82271186440678e-06,
      "loss": 0.2234,
      "step": 1025
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.814237288135594e-06,
      "loss": 0.1974,
      "step": 1050
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.805762711864407e-06,
      "loss": 0.2175,
      "step": 1075
    },
    {
      "epoch": 0.38,
      "learning_rate": 9.797288135593221e-06,
      "loss": 0.207,
      "step": 1100
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.788813559322034e-06,
      "loss": 0.1922,
      "step": 1125
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.780338983050847e-06,
      "loss": 0.2128,
      "step": 1150
    },
    {
      "epoch": 0.4,
      "learning_rate": 9.771864406779662e-06,
      "loss": 0.2038,
      "step": 1175
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.763389830508475e-06,
      "loss": 0.1996,
      "step": 1200
    },
    {
      "epoch": 0.42,
      "learning_rate": 9.75491525423729e-06,
      "loss": 0.1982,
      "step": 1225
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.746440677966103e-06,
      "loss": 0.2026,
      "step": 1250
    },
    {
      "epoch": 0.44,
      "learning_rate": 9.737966101694916e-06,
      "loss": 0.2056,
      "step": 1275
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.72949152542373e-06,
      "loss": 0.2038,
      "step": 1300
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.721016949152543e-06,
      "loss": 0.1906,
      "step": 1325
    },
    {
      "epoch": 0.46,
      "learning_rate": 9.712542372881358e-06,
      "loss": 0.1953,
      "step": 1350
    },
    {
      "epoch": 0.47,
      "learning_rate": 9.70406779661017e-06,
      "loss": 0.1952,
      "step": 1375
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.695593220338984e-06,
      "loss": 0.1988,
      "step": 1400
    },
    {
      "epoch": 0.49,
      "learning_rate": 9.687118644067798e-06,
      "loss": 0.1956,
      "step": 1425
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.678644067796611e-06,
      "loss": 0.19,
      "step": 1450
    },
    {
      "epoch": 0.51,
      "learning_rate": 9.670169491525424e-06,
      "loss": 0.2183,
      "step": 1475
    },
    {
      "epoch": 0.51,
      "learning_rate": 9.661694915254239e-06,
      "loss": 0.1744,
      "step": 1500
    },
    {
      "epoch": 0.51,
      "eval_loss": 0.1370771825313568,
      "eval_runtime": 14137.0887,
      "eval_samples_per_second": 1.582,
      "eval_steps_per_second": 0.066,
      "eval_wer": 4.992967013022369,
      "step": 1500
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.653220338983052e-06,
      "loss": 0.2261,
      "step": 1525
    },
    {
      "epoch": 0.53,
      "learning_rate": 9.644745762711865e-06,
      "loss": 0.2001,
      "step": 1550
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.636271186440678e-06,
      "loss": 0.1984,
      "step": 1575
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.627796610169493e-06,
      "loss": 0.2003,
      "step": 1600
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.619322033898306e-06,
      "loss": 0.1771,
      "step": 1625
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.610847457627119e-06,
      "loss": 0.1979,
      "step": 1650
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.602372881355932e-06,
      "loss": 0.2013,
      "step": 1675
    },
    {
      "epoch": 0.58,
      "learning_rate": 9.593898305084746e-06,
      "loss": 0.1899,
      "step": 1700
    },
    {
      "epoch": 0.59,
      "learning_rate": 9.58542372881356e-06,
      "loss": 0.1782,
      "step": 1725
    },
    {
      "epoch": 0.6,
      "learning_rate": 9.576949152542374e-06,
      "loss": 0.1689,
      "step": 1750
    },
    {
      "epoch": 0.61,
      "learning_rate": 9.568474576271187e-06,
      "loss": 0.1985,
      "step": 1775
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.56e-06,
      "loss": 0.1788,
      "step": 1800
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.551525423728815e-06,
      "loss": 0.186,
      "step": 1825
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.543050847457628e-06,
      "loss": 0.1873,
      "step": 1850
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.53457627118644e-06,
      "loss": 0.1944,
      "step": 1875
    },
    {
      "epoch": 0.65,
      "learning_rate": 9.526101694915255e-06,
      "loss": 0.1954,
      "step": 1900
    },
    {
      "epoch": 0.66,
      "learning_rate": 9.517627118644068e-06,
      "loss": 0.1948,
      "step": 1925
    },
    {
      "epoch": 0.67,
      "learning_rate": 9.509152542372883e-06,
      "loss": 0.1785,
      "step": 1950
    },
    {
      "epoch": 0.68,
      "learning_rate": 9.500677966101696e-06,
      "loss": 0.1795,
      "step": 1975
    },
    {
      "epoch": 0.69,
      "learning_rate": 9.492203389830509e-06,
      "loss": 0.1821,
      "step": 2000
    },
    {
      "epoch": 0.69,
      "eval_loss": 0.13245970010757446,
      "eval_runtime": 13962.1433,
      "eval_samples_per_second": 1.602,
      "eval_steps_per_second": 0.067,
      "eval_wer": 4.777893733835473,
      "step": 2000
    },
    {
      "epoch": 0.69,
      "learning_rate": 9.483728813559324e-06,
      "loss": 0.1895,
      "step": 2025
    },
    {
      "epoch": 0.7,
      "learning_rate": 9.475254237288137e-06,
      "loss": 0.1693,
      "step": 2050
    },
    {
      "epoch": 0.71,
      "learning_rate": 9.466779661016951e-06,
      "loss": 0.1791,
      "step": 2075
    },
    {
      "epoch": 0.72,
      "learning_rate": 9.458305084745764e-06,
      "loss": 0.1889,
      "step": 2100
    },
    {
      "epoch": 0.73,
      "learning_rate": 9.449830508474577e-06,
      "loss": 0.17,
      "step": 2125
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.44135593220339e-06,
      "loss": 0.2053,
      "step": 2150
    },
    {
      "epoch": 0.75,
      "learning_rate": 9.432881355932205e-06,
      "loss": 0.1754,
      "step": 2175
    },
    {
      "epoch": 0.75,
      "learning_rate": 9.424406779661018e-06,
      "loss": 0.1791,
      "step": 2200
    },
    {
      "epoch": 0.76,
      "learning_rate": 9.41593220338983e-06,
      "loss": 0.1901,
      "step": 2225
    },
    {
      "epoch": 0.77,
      "learning_rate": 9.407457627118645e-06,
      "loss": 0.1682,
      "step": 2250
    },
    {
      "epoch": 0.78,
      "learning_rate": 9.398983050847458e-06,
      "loss": 0.176,
      "step": 2275
    },
    {
      "epoch": 0.79,
      "learning_rate": 9.390508474576271e-06,
      "loss": 0.1846,
      "step": 2300
    },
    {
      "epoch": 0.8,
      "learning_rate": 9.382033898305084e-06,
      "loss": 0.1789,
      "step": 2325
    },
    {
      "epoch": 0.81,
      "learning_rate": 9.373559322033899e-06,
      "loss": 0.1645,
      "step": 2350
    },
    {
      "epoch": 0.82,
      "learning_rate": 9.365084745762712e-06,
      "loss": 0.1772,
      "step": 2375
    },
    {
      "epoch": 0.82,
      "learning_rate": 9.356610169491525e-06,
      "loss": 0.1652,
      "step": 2400
    },
    {
      "epoch": 0.83,
      "learning_rate": 9.34813559322034e-06,
      "loss": 0.1752,
      "step": 2425
    },
    {
      "epoch": 0.84,
      "learning_rate": 9.339661016949153e-06,
      "loss": 0.1864,
      "step": 2450
    },
    {
      "epoch": 0.85,
      "learning_rate": 9.331186440677967e-06,
      "loss": 0.17,
      "step": 2475
    },
    {
      "epoch": 0.86,
      "learning_rate": 9.32271186440678e-06,
      "loss": 0.1705,
      "step": 2500
    },
    {
      "epoch": 0.86,
      "eval_loss": 0.13045215606689453,
      "eval_runtime": 14199.68,
      "eval_samples_per_second": 1.575,
      "eval_steps_per_second": 0.066,
      "eval_wer": 4.807840646127319,
      "step": 2500
    },
    {
      "epoch": 0.87,
      "learning_rate": 9.314237288135593e-06,
      "loss": 0.1801,
      "step": 2525
    },
    {
      "epoch": 0.88,
      "learning_rate": 9.305762711864408e-06,
      "loss": 0.1708,
      "step": 2550
    },
    {
      "epoch": 0.88,
      "learning_rate": 9.297288135593221e-06,
      "loss": 0.1831,
      "step": 2575
    },
    {
      "epoch": 0.89,
      "learning_rate": 9.288813559322036e-06,
      "loss": 0.1803,
      "step": 2600
    },
    {
      "epoch": 0.9,
      "learning_rate": 9.280338983050849e-06,
      "loss": 0.1787,
      "step": 2625
    },
    {
      "epoch": 0.91,
      "learning_rate": 9.271864406779662e-06,
      "loss": 0.1561,
      "step": 2650
    },
    {
      "epoch": 0.92,
      "learning_rate": 9.263389830508476e-06,
      "loss": 0.1632,
      "step": 2675
    },
    {
      "epoch": 0.93,
      "learning_rate": 9.25491525423729e-06,
      "loss": 0.1867,
      "step": 2700
    },
    {
      "epoch": 0.94,
      "learning_rate": 9.246440677966102e-06,
      "loss": 0.1767,
      "step": 2725
    },
    {
      "epoch": 0.94,
      "learning_rate": 9.237966101694917e-06,
      "loss": 0.1662,
      "step": 2750
    },
    {
      "epoch": 0.95,
      "learning_rate": 9.22949152542373e-06,
      "loss": 0.179,
      "step": 2775
    },
    {
      "epoch": 0.96,
      "learning_rate": 9.221016949152543e-06,
      "loss": 0.1696,
      "step": 2800
    },
    {
      "epoch": 0.97,
      "learning_rate": 9.212542372881358e-06,
      "loss": 0.1697,
      "step": 2825
    },
    {
      "epoch": 0.98,
      "learning_rate": 9.20406779661017e-06,
      "loss": 0.1776,
      "step": 2850
    },
    {
      "epoch": 0.99,
      "learning_rate": 9.195593220338984e-06,
      "loss": 0.1624,
      "step": 2875
    },
    {
      "epoch": 1.0,
      "learning_rate": 9.187118644067796e-06,
      "loss": 0.1629,
      "step": 2900
    },
    {
      "epoch": 1.0,
      "learning_rate": 9.178644067796611e-06,
      "loss": 0.1614,
      "step": 2925
    },
    {
      "epoch": 1.01,
      "learning_rate": 9.170169491525424e-06,
      "loss": 0.1087,
      "step": 2950
    },
    {
      "epoch": 1.02,
      "learning_rate": 9.161694915254237e-06,
      "loss": 0.1122,
      "step": 2975
    },
    {
      "epoch": 1.03,
      "learning_rate": 9.153220338983052e-06,
      "loss": 0.1069,
      "step": 3000
    },
    {
      "epoch": 1.03,
      "eval_loss": 0.1263168603181839,
      "eval_runtime": 13889.4582,
      "eval_samples_per_second": 1.61,
      "eval_steps_per_second": 0.067,
      "eval_wer": 4.3962974726620985,
      "step": 3000
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.144745762711865e-06,
      "loss": 0.1051,
      "step": 3025
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.136271186440678e-06,
      "loss": 0.103,
      "step": 3050
    },
    {
      "epoch": 1.06,
      "learning_rate": 9.127796610169492e-06,
      "loss": 0.1039,
      "step": 3075
    },
    {
      "epoch": 1.06,
      "learning_rate": 9.119322033898305e-06,
      "loss": 0.1113,
      "step": 3100
    },
    {
      "epoch": 1.07,
      "learning_rate": 9.110847457627118e-06,
      "loss": 0.1092,
      "step": 3125
    },
    {
      "epoch": 1.08,
      "learning_rate": 9.102372881355933e-06,
      "loss": 0.1272,
      "step": 3150
    },
    {
      "epoch": 1.09,
      "learning_rate": 9.093898305084746e-06,
      "loss": 0.1079,
      "step": 3175
    },
    {
      "epoch": 1.1,
      "learning_rate": 9.08542372881356e-06,
      "loss": 0.1137,
      "step": 3200
    },
    {
      "epoch": 1.11,
      "learning_rate": 9.076949152542374e-06,
      "loss": 0.1097,
      "step": 3225
    },
    {
      "epoch": 1.12,
      "learning_rate": 9.068474576271187e-06,
      "loss": 0.1127,
      "step": 3250
    },
    {
      "epoch": 1.12,
      "learning_rate": 9.060000000000001e-06,
      "loss": 0.1139,
      "step": 3275
    },
    {
      "epoch": 1.13,
      "learning_rate": 9.051525423728814e-06,
      "loss": 0.1059,
      "step": 3300
    },
    {
      "epoch": 1.14,
      "learning_rate": 9.043050847457629e-06,
      "loss": 0.1103,
      "step": 3325
    },
    {
      "epoch": 1.15,
      "learning_rate": 9.034576271186442e-06,
      "loss": 0.1238,
      "step": 3350
    },
    {
      "epoch": 1.16,
      "learning_rate": 9.026101694915255e-06,
      "loss": 0.1008,
      "step": 3375
    },
    {
      "epoch": 1.17,
      "learning_rate": 9.01762711864407e-06,
      "loss": 0.1077,
      "step": 3400
    },
    {
      "epoch": 1.18,
      "learning_rate": 9.009152542372883e-06,
      "loss": 0.1086,
      "step": 3425
    },
    {
      "epoch": 1.18,
      "learning_rate": 9.000677966101696e-06,
      "loss": 0.1126,
      "step": 3450
    },
    {
      "epoch": 1.19,
      "learning_rate": 8.99220338983051e-06,
      "loss": 0.1151,
      "step": 3475
    },
    {
      "epoch": 1.2,
      "learning_rate": 8.983728813559323e-06,
      "loss": 0.1098,
      "step": 3500
    },
    {
      "epoch": 1.2,
      "eval_loss": 0.12618736922740936,
      "eval_runtime": 14301.962,
      "eval_samples_per_second": 1.564,
      "eval_steps_per_second": 0.065,
      "eval_wer": 4.478878352012342,
      "step": 3500
    },
    {
      "epoch": 1.21,
      "learning_rate": 8.975254237288136e-06,
      "loss": 0.1181,
      "step": 3525
    },
    {
      "epoch": 1.22,
      "learning_rate": 8.96677966101695e-06,
      "loss": 0.1102,
      "step": 3550
    },
    {
      "epoch": 1.23,
      "learning_rate": 8.958305084745764e-06,
      "loss": 0.1046,
      "step": 3575
    },
    {
      "epoch": 1.24,
      "learning_rate": 8.949830508474577e-06,
      "loss": 0.1146,
      "step": 3600
    },
    {
      "epoch": 1.24,
      "learning_rate": 8.94135593220339e-06,
      "loss": 0.1192,
      "step": 3625
    },
    {
      "epoch": 1.25,
      "learning_rate": 8.932881355932203e-06,
      "loss": 0.1087,
      "step": 3650
    },
    {
      "epoch": 1.26,
      "learning_rate": 8.924406779661017e-06,
      "loss": 0.1143,
      "step": 3675
    },
    {
      "epoch": 1.27,
      "learning_rate": 8.91593220338983e-06,
      "loss": 0.1039,
      "step": 3700
    },
    {
      "epoch": 1.28,
      "learning_rate": 8.907457627118645e-06,
      "loss": 0.1194,
      "step": 3725
    },
    {
      "epoch": 1.29,
      "learning_rate": 8.898983050847458e-06,
      "loss": 0.1128,
      "step": 3750
    },
    {
      "epoch": 1.3,
      "learning_rate": 8.890508474576271e-06,
      "loss": 0.108,
      "step": 3775
    },
    {
      "epoch": 1.3,
      "learning_rate": 8.882033898305086e-06,
      "loss": 0.1049,
      "step": 3800
    },
    {
      "epoch": 1.31,
      "learning_rate": 8.873559322033899e-06,
      "loss": 0.1253,
      "step": 3825
    },
    {
      "epoch": 1.32,
      "learning_rate": 8.865084745762712e-06,
      "loss": 0.1165,
      "step": 3850
    },
    {
      "epoch": 1.33,
      "learning_rate": 8.856610169491526e-06,
      "loss": 0.1189,
      "step": 3875
    },
    {
      "epoch": 1.34,
      "learning_rate": 8.84813559322034e-06,
      "loss": 0.1115,
      "step": 3900
    },
    {
      "epoch": 1.35,
      "learning_rate": 8.839661016949154e-06,
      "loss": 0.1214,
      "step": 3925
    },
    {
      "epoch": 1.36,
      "learning_rate": 8.831186440677967e-06,
      "loss": 0.101,
      "step": 3950
    },
    {
      "epoch": 1.36,
      "learning_rate": 8.82271186440678e-06,
      "loss": 0.1263,
      "step": 3975
    },
    {
      "epoch": 1.37,
      "learning_rate": 8.814237288135595e-06,
      "loss": 0.1059,
      "step": 4000
    },
    {
      "epoch": 1.37,
      "eval_loss": 0.12528666853904724,
      "eval_runtime": 14038.6098,
      "eval_samples_per_second": 1.593,
      "eval_steps_per_second": 0.066,
      "eval_wer": 4.441217841099868,
      "step": 4000
    },
    {
      "epoch": 1.38,
      "learning_rate": 8.805762711864408e-06,
      "loss": 0.1241,
      "step": 4025
    },
    {
      "epoch": 1.39,
      "learning_rate": 8.797288135593222e-06,
      "loss": 0.1159,
      "step": 4050
    },
    {
      "epoch": 1.4,
      "learning_rate": 8.788813559322035e-06,
      "loss": 0.1161,
      "step": 4075
    },
    {
      "epoch": 1.41,
      "learning_rate": 8.780338983050848e-06,
      "loss": 0.1064,
      "step": 4100
    },
    {
      "epoch": 1.42,
      "learning_rate": 8.771864406779661e-06,
      "loss": 0.1106,
      "step": 4125
    },
    {
      "epoch": 1.42,
      "learning_rate": 8.763389830508476e-06,
      "loss": 0.1048,
      "step": 4150
    },
    {
      "epoch": 1.43,
      "learning_rate": 8.754915254237289e-06,
      "loss": 0.1122,
      "step": 4175
    },
    {
      "epoch": 1.44,
      "learning_rate": 8.746440677966102e-06,
      "loss": 0.1262,
      "step": 4200
    },
    {
      "epoch": 1.45,
      "learning_rate": 8.737966101694917e-06,
      "loss": 0.1277,
      "step": 4225
    },
    {
      "epoch": 1.46,
      "learning_rate": 8.72949152542373e-06,
      "loss": 0.112,
      "step": 4250
    },
    {
      "epoch": 1.47,
      "learning_rate": 8.721016949152543e-06,
      "loss": 0.1011,
      "step": 4275
    },
    {
      "epoch": 1.48,
      "learning_rate": 8.712542372881356e-06,
      "loss": 0.106,
      "step": 4300
    },
    {
      "epoch": 1.48,
      "learning_rate": 8.70406779661017e-06,
      "loss": 0.1206,
      "step": 4325
    },
    {
      "epoch": 1.49,
      "learning_rate": 8.695593220338983e-06,
      "loss": 0.1097,
      "step": 4350
    },
    {
      "epoch": 1.5,
      "learning_rate": 8.687118644067796e-06,
      "loss": 0.1212,
      "step": 4375
    },
    {
      "epoch": 1.51,
      "learning_rate": 8.678644067796611e-06,
      "loss": 0.1068,
      "step": 4400
    },
    {
      "epoch": 1.52,
      "learning_rate": 8.670169491525424e-06,
      "loss": 0.1096,
      "step": 4425
    },
    {
      "epoch": 1.53,
      "learning_rate": 8.661694915254239e-06,
      "loss": 0.1062,
      "step": 4450
    },
    {
      "epoch": 1.54,
      "learning_rate": 8.653220338983051e-06,
      "loss": 0.1,
      "step": 4475
    },
    {
      "epoch": 1.54,
      "learning_rate": 8.644745762711864e-06,
      "loss": 0.1088,
      "step": 4500
    },
    {
      "epoch": 1.54,
      "eval_loss": 0.12188256531953812,
      "eval_runtime": 13978.7976,
      "eval_samples_per_second": 1.6,
      "eval_steps_per_second": 0.067,
      "eval_wer": 4.305549253595898,
      "step": 4500
    },
    {
      "epoch": 1.55,
      "learning_rate": 8.636271186440679e-06,
      "loss": 0.122,
      "step": 4525
    },
    {
      "epoch": 1.56,
      "learning_rate": 8.627796610169492e-06,
      "loss": 0.1094,
      "step": 4550
    },
    {
      "epoch": 1.57,
      "learning_rate": 8.619322033898305e-06,
      "loss": 0.1163,
      "step": 4575
    },
    {
      "epoch": 1.58,
      "learning_rate": 8.61084745762712e-06,
      "loss": 0.127,
      "step": 4600
    },
    {
      "epoch": 1.59,
      "learning_rate": 8.602372881355933e-06,
      "loss": 0.1122,
      "step": 4625
    },
    {
      "epoch": 1.6,
      "learning_rate": 8.593898305084747e-06,
      "loss": 0.1056,
      "step": 4650
    },
    {
      "epoch": 1.6,
      "learning_rate": 8.58542372881356e-06,
      "loss": 0.1051,
      "step": 4675
    },
    {
      "epoch": 1.61,
      "learning_rate": 8.576949152542373e-06,
      "loss": 0.1087,
      "step": 4700
    },
    {
      "epoch": 1.62,
      "learning_rate": 8.568474576271188e-06,
      "loss": 0.1155,
      "step": 4725
    },
    {
      "epoch": 1.63,
      "learning_rate": 8.560000000000001e-06,
      "loss": 0.114,
      "step": 4750
    },
    {
      "epoch": 1.64,
      "learning_rate": 8.551525423728814e-06,
      "loss": 0.1286,
      "step": 4775
    },
    {
      "epoch": 1.65,
      "learning_rate": 8.543050847457629e-06,
      "loss": 0.1052,
      "step": 4800
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.534576271186442e-06,
      "loss": 0.1156,
      "step": 4825
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.526101694915255e-06,
      "loss": 0.0998,
      "step": 4850
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.517627118644068e-06,
      "loss": 0.1147,
      "step": 4875
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.509152542372882e-06,
      "loss": 0.1007,
      "step": 4900
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.500677966101695e-06,
      "loss": 0.1102,
      "step": 4925
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.492203389830508e-06,
      "loss": 0.1113,
      "step": 4950
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.483728813559323e-06,
      "loss": 0.1118,
      "step": 4975
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.475254237288136e-06,
      "loss": 0.1074,
      "step": 5000
    },
    {
      "epoch": 1.72,
      "eval_loss": 0.12055900692939758,
      "eval_runtime": 13950.039,
      "eval_samples_per_second": 1.603,
      "eval_steps_per_second": 0.067,
      "eval_wer": 4.194382685239803,
      "step": 5000
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.466779661016949e-06,
      "loss": 0.1024,
      "step": 5025
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.458305084745764e-06,
      "loss": 0.1078,
      "step": 5050
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.449830508474577e-06,
      "loss": 0.1198,
      "step": 5075
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.44135593220339e-06,
      "loss": 0.1052,
      "step": 5100
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.432881355932204e-06,
      "loss": 0.1104,
      "step": 5125
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.424406779661017e-06,
      "loss": 0.111,
      "step": 5150
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.415932203389832e-06,
      "loss": 0.114,
      "step": 5175
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.407457627118645e-06,
      "loss": 0.1074,
      "step": 5200
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.398983050847458e-06,
      "loss": 0.1117,
      "step": 5225
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.390508474576273e-06,
      "loss": 0.1059,
      "step": 5250
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.382033898305085e-06,
      "loss": 0.126,
      "step": 5275
    },
    {
      "epoch": 1.82,
      "learning_rate": 8.3735593220339e-06,
      "loss": 0.1037,
      "step": 5300
    },
    {
      "epoch": 1.83,
      "learning_rate": 8.365084745762713e-06,
      "loss": 0.0988,
      "step": 5325
    },
    {
      "epoch": 1.84,
      "learning_rate": 8.356610169491526e-06,
      "loss": 0.1053,
      "step": 5350
    },
    {
      "epoch": 1.84,
      "learning_rate": 8.34813559322034e-06,
      "loss": 0.1128,
      "step": 5375
    },
    {
      "epoch": 1.85,
      "learning_rate": 8.339661016949154e-06,
      "loss": 0.1132,
      "step": 5400
    },
    {
      "epoch": 1.86,
      "learning_rate": 8.331186440677967e-06,
      "loss": 0.1083,
      "step": 5425
    },
    {
      "epoch": 1.87,
      "learning_rate": 8.32271186440678e-06,
      "loss": 0.128,
      "step": 5450
    },
    {
      "epoch": 1.88,
      "learning_rate": 8.314237288135594e-06,
      "loss": 0.1044,
      "step": 5475
    },
    {
      "epoch": 1.89,
      "learning_rate": 8.305762711864407e-06,
      "loss": 0.1037,
      "step": 5500
    },
    {
      "epoch": 1.89,
      "eval_loss": 0.1189337745308876,
      "eval_runtime": 14041.5407,
      "eval_samples_per_second": 1.593,
      "eval_steps_per_second": 0.066,
      "eval_wer": 4.197558872907119,
      "step": 5500
    },
    {
      "epoch": 1.9,
      "learning_rate": 8.29728813559322e-06,
      "loss": 0.1185,
      "step": 5525
    },
    {
      "epoch": 1.9,
      "learning_rate": 8.288813559322035e-06,
      "loss": 0.1117,
      "step": 5550
    },
    {
      "epoch": 1.91,
      "learning_rate": 8.280338983050848e-06,
      "loss": 0.1185,
      "step": 5575
    },
    {
      "epoch": 1.92,
      "learning_rate": 8.271864406779661e-06,
      "loss": 0.1155,
      "step": 5600
    },
    {
      "epoch": 1.93,
      "learning_rate": 8.263389830508474e-06,
      "loss": 0.1203,
      "step": 5625
    },
    {
      "epoch": 1.94,
      "learning_rate": 8.254915254237289e-06,
      "loss": 0.1065,
      "step": 5650
    },
    {
      "epoch": 1.95,
      "learning_rate": 8.246440677966102e-06,
      "loss": 0.1063,
      "step": 5675
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.237966101694916e-06,
      "loss": 0.113,
      "step": 5700
    },
    {
      "epoch": 1.96,
      "learning_rate": 8.22949152542373e-06,
      "loss": 0.1073,
      "step": 5725
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.221016949152542e-06,
      "loss": 0.1045,
      "step": 5750
    },
    {
      "epoch": 1.98,
      "learning_rate": 8.212542372881357e-06,
      "loss": 0.1053,
      "step": 5775
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.20406779661017e-06,
      "loss": 0.0961,
      "step": 5800
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.195593220338983e-06,
      "loss": 0.1045,
      "step": 5825
    },
    {
      "epoch": 2.01,
      "learning_rate": 8.187118644067798e-06,
      "loss": 0.0678,
      "step": 5850
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.17864406779661e-06,
      "loss": 0.0583,
      "step": 5875
    },
    {
      "epoch": 2.02,
      "learning_rate": 8.170169491525425e-06,
      "loss": 0.0615,
      "step": 5900
    },
    {
      "epoch": 2.03,
      "learning_rate": 8.161694915254238e-06,
      "loss": 0.0624,
      "step": 5925
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.153220338983051e-06,
      "loss": 0.0561,
      "step": 5950
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.144745762711866e-06,
      "loss": 0.0633,
      "step": 5975
    },
    {
      "epoch": 2.06,
      "learning_rate": 8.136271186440679e-06,
      "loss": 0.0575,
      "step": 6000
    },
    {
      "epoch": 2.06,
      "eval_loss": 0.12541243433952332,
      "eval_runtime": 13932.2659,
      "eval_samples_per_second": 1.605,
      "eval_steps_per_second": 0.067,
      "eval_wer": 4.1412949770860745,
      "step": 6000
    },
    {
      "epoch": 2.07,
      "learning_rate": 8.127796610169494e-06,
      "loss": 0.0622,
      "step": 6025
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.119322033898307e-06,
      "loss": 0.0632,
      "step": 6050
    },
    {
      "epoch": 2.08,
      "learning_rate": 8.11084745762712e-06,
      "loss": 0.0542,
      "step": 6075
    },
    {
      "epoch": 2.09,
      "learning_rate": 8.102372881355932e-06,
      "loss": 0.0644,
      "step": 6100
    },
    {
      "epoch": 2.1,
      "learning_rate": 8.093898305084747e-06,
      "loss": 0.0671,
      "step": 6125
    },
    {
      "epoch": 2.11,
      "learning_rate": 8.08542372881356e-06,
      "loss": 0.0532,
      "step": 6150
    },
    {
      "epoch": 2.12,
      "learning_rate": 8.076949152542373e-06,
      "loss": 0.064,
      "step": 6175
    },
    {
      "epoch": 2.13,
      "learning_rate": 8.068474576271186e-06,
      "loss": 0.0624,
      "step": 6200
    },
    {
      "epoch": 2.14,
      "learning_rate": 8.06e-06,
      "loss": 0.0722,
      "step": 6225
    },
    {
      "epoch": 2.14,
      "learning_rate": 8.051525423728814e-06,
      "loss": 0.0674,
      "step": 6250
    },
    {
      "epoch": 2.15,
      "learning_rate": 8.043050847457627e-06,
      "loss": 0.0573,
      "step": 6275
    },
    {
      "epoch": 2.16,
      "learning_rate": 8.034576271186441e-06,
      "loss": 0.0618,
      "step": 6300
    },
    {
      "epoch": 2.17,
      "learning_rate": 8.026101694915254e-06,
      "loss": 0.0647,
      "step": 6325
    },
    {
      "epoch": 2.18,
      "learning_rate": 8.017627118644067e-06,
      "loss": 0.0667,
      "step": 6350
    },
    {
      "epoch": 2.19,
      "learning_rate": 8.009152542372882e-06,
      "loss": 0.0569,
      "step": 6375
    },
    {
      "epoch": 2.2,
      "learning_rate": 8.000677966101695e-06,
      "loss": 0.0755,
      "step": 6400
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.99220338983051e-06,
      "loss": 0.0617,
      "step": 6425
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.983728813559323e-06,
      "loss": 0.0592,
      "step": 6450
    },
    {
      "epoch": 2.22,
      "learning_rate": 7.975254237288136e-06,
      "loss": 0.0578,
      "step": 6475
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.96677966101695e-06,
      "loss": 0.0623,
      "step": 6500
    },
    {
      "epoch": 2.23,
      "eval_loss": 0.12663590908050537,
      "eval_runtime": 14047.6478,
      "eval_samples_per_second": 1.592,
      "eval_steps_per_second": 0.066,
      "eval_wer": 4.139933753800082,
      "step": 6500
    },
    {
      "epoch": 2.24,
      "learning_rate": 7.958305084745763e-06,
      "loss": 0.0718,
      "step": 6525
    },
    {
      "epoch": 2.25,
      "learning_rate": 7.949830508474576e-06,
      "loss": 0.0643,
      "step": 6550
    },
    {
      "epoch": 2.26,
      "learning_rate": 7.941355932203391e-06,
      "loss": 0.0648,
      "step": 6575
    },
    {
      "epoch": 2.26,
      "learning_rate": 7.932881355932204e-06,
      "loss": 0.0632,
      "step": 6600
    },
    {
      "epoch": 2.27,
      "learning_rate": 7.924406779661019e-06,
      "loss": 0.0686,
      "step": 6625
    },
    {
      "epoch": 2.28,
      "learning_rate": 7.915932203389832e-06,
      "loss": 0.0675,
      "step": 6650
    },
    {
      "epoch": 2.29,
      "learning_rate": 7.907457627118645e-06,
      "loss": 0.0624,
      "step": 6675
    },
    {
      "epoch": 2.3,
      "learning_rate": 7.89898305084746e-06,
      "loss": 0.0553,
      "step": 6700
    },
    {
      "epoch": 2.31,
      "learning_rate": 7.890508474576272e-06,
      "loss": 0.0643,
      "step": 6725
    },
    {
      "epoch": 2.32,
      "learning_rate": 7.882033898305085e-06,
      "loss": 0.0573,
      "step": 6750
    },
    {
      "epoch": 2.32,
      "learning_rate": 7.8735593220339e-06,
      "loss": 0.0698,
      "step": 6775
    },
    {
      "epoch": 2.33,
      "learning_rate": 7.865084745762713e-06,
      "loss": 0.0659,
      "step": 6800
    },
    {
      "epoch": 2.34,
      "learning_rate": 7.856610169491526e-06,
      "loss": 0.0654,
      "step": 6825
    },
    {
      "epoch": 2.35,
      "learning_rate": 7.848135593220339e-06,
      "loss": 0.064,
      "step": 6850
    },
    {
      "epoch": 2.36,
      "learning_rate": 7.839661016949153e-06,
      "loss": 0.0615,
      "step": 6875
    },
    {
      "epoch": 2.37,
      "learning_rate": 7.831186440677966e-06,
      "loss": 0.0579,
      "step": 6900
    },
    {
      "epoch": 2.38,
      "learning_rate": 7.82271186440678e-06,
      "loss": 0.0675,
      "step": 6925
    },
    {
      "epoch": 2.39,
      "learning_rate": 7.814237288135594e-06,
      "loss": 0.0682,
      "step": 6950
    },
    {
      "epoch": 2.39,
      "learning_rate": 7.805762711864407e-06,
      "loss": 0.0682,
      "step": 6975
    },
    {
      "epoch": 2.4,
      "learning_rate": 7.79728813559322e-06,
      "loss": 0.0745,
      "step": 7000
    },
    {
      "epoch": 2.4,
      "eval_loss": 0.12451925128698349,
      "eval_runtime": 13911.3227,
      "eval_samples_per_second": 1.608,
      "eval_steps_per_second": 0.067,
      "eval_wer": 4.105449430554926,
      "step": 7000
    },
    {
      "epoch": 2.41,
      "learning_rate": 7.788813559322035e-06,
      "loss": 0.0574,
      "step": 7025
    },
    {
      "epoch": 2.42,
      "learning_rate": 7.780338983050848e-06,
      "loss": 0.0597,
      "step": 7050
    },
    {
      "epoch": 2.43,
      "learning_rate": 7.77186440677966e-06,
      "loss": 0.0613,
      "step": 7075
    },
    {
      "epoch": 2.44,
      "learning_rate": 7.763389830508475e-06,
      "loss": 0.0712,
      "step": 7100
    },
    {
      "epoch": 2.45,
      "learning_rate": 7.754915254237288e-06,
      "loss": 0.0622,
      "step": 7125
    },
    {
      "epoch": 2.45,
      "learning_rate": 7.746440677966103e-06,
      "loss": 0.0631,
      "step": 7150
    },
    {
      "epoch": 2.46,
      "learning_rate": 7.737966101694916e-06,
      "loss": 0.0696,
      "step": 7175
    },
    {
      "epoch": 2.47,
      "learning_rate": 7.729491525423729e-06,
      "loss": 0.0657,
      "step": 7200
    },
    {
      "epoch": 2.48,
      "learning_rate": 7.721016949152544e-06,
      "loss": 0.0593,
      "step": 7225
    },
    {
      "epoch": 2.49,
      "learning_rate": 7.712542372881357e-06,
      "loss": 0.0644,
      "step": 7250
    },
    {
      "epoch": 2.5,
      "learning_rate": 7.70406779661017e-06,
      "loss": 0.0602,
      "step": 7275
    },
    {
      "epoch": 2.51,
      "learning_rate": 7.695593220338984e-06,
      "loss": 0.0662,
      "step": 7300
    },
    {
      "epoch": 2.51,
      "learning_rate": 7.687118644067797e-06,
      "loss": 0.0654,
      "step": 7325
    },
    {
      "epoch": 2.52,
      "learning_rate": 7.678983050847458e-06,
      "loss": 0.0699,
      "step": 7350
    },
    {
      "epoch": 2.53,
      "learning_rate": 7.670508474576271e-06,
      "loss": 0.0657,
      "step": 7375
    },
    {
      "epoch": 2.54,
      "learning_rate": 7.662033898305086e-06,
      "loss": 0.0668,
      "step": 7400
    },
    {
      "epoch": 2.55,
      "learning_rate": 7.653559322033899e-06,
      "loss": 0.0598,
      "step": 7425
    },
    {
      "epoch": 2.56,
      "learning_rate": 7.645084745762712e-06,
      "loss": 0.062,
      "step": 7450
    },
    {
      "epoch": 2.57,
      "learning_rate": 7.636610169491526e-06,
      "loss": 0.0657,
      "step": 7475
    },
    {
      "epoch": 2.57,
      "learning_rate": 7.628135593220339e-06,
      "loss": 0.0686,
      "step": 7500
    },
    {
      "epoch": 2.57,
      "eval_loss": 0.12511542439460754,
      "eval_runtime": 13875.855,
      "eval_samples_per_second": 1.612,
      "eval_steps_per_second": 0.067,
      "eval_wer": 4.054176686782522,
      "step": 7500
    }
  ],
  "max_steps": 30000,
  "num_train_epochs": 11,
  "total_flos": 3.6740551944830976e+20,
  "trial_name": null,
  "trial_params": null
}