anton-l's picture
anton-l HF staff
End of training
ff8617c
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 30.0,
"global_step": 1200,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 0.0,
"loss": 16.7868,
"step": 1
},
{
"epoch": 0.05,
"learning_rate": 0.0,
"loss": 14.8595,
"step": 2
},
{
"epoch": 0.07,
"learning_rate": 1.4999999999999998e-06,
"loss": 13.3845,
"step": 3
},
{
"epoch": 0.1,
"learning_rate": 2.9999999999999997e-06,
"loss": 13.1354,
"step": 4
},
{
"epoch": 0.12,
"learning_rate": 4.499999999999999e-06,
"loss": 13.751,
"step": 5
},
{
"epoch": 0.15,
"learning_rate": 5.999999999999999e-06,
"loss": 12.7844,
"step": 6
},
{
"epoch": 0.17,
"learning_rate": 7.499999999999999e-06,
"loss": 16.4185,
"step": 7
},
{
"epoch": 0.2,
"learning_rate": 8.999999999999999e-06,
"loss": 14.9368,
"step": 8
},
{
"epoch": 0.23,
"learning_rate": 1.05e-05,
"loss": 13.8337,
"step": 9
},
{
"epoch": 0.25,
"learning_rate": 1.1999999999999999e-05,
"loss": 13.0948,
"step": 10
},
{
"epoch": 0.28,
"learning_rate": 1.3499999999999998e-05,
"loss": 13.1814,
"step": 11
},
{
"epoch": 0.3,
"learning_rate": 1.4999999999999999e-05,
"loss": 12.3933,
"step": 12
},
{
"epoch": 0.33,
"learning_rate": 1.6499999999999998e-05,
"loss": 14.1081,
"step": 13
},
{
"epoch": 0.35,
"learning_rate": 1.7999999999999997e-05,
"loss": 14.3403,
"step": 14
},
{
"epoch": 0.38,
"learning_rate": 1.95e-05,
"loss": 14.1041,
"step": 15
},
{
"epoch": 0.4,
"learning_rate": 2.1e-05,
"loss": 12.2322,
"step": 16
},
{
"epoch": 0.42,
"learning_rate": 2.2499999999999998e-05,
"loss": 12.5879,
"step": 17
},
{
"epoch": 0.45,
"learning_rate": 2.3999999999999997e-05,
"loss": 12.9614,
"step": 18
},
{
"epoch": 0.47,
"learning_rate": 2.55e-05,
"loss": 12.602,
"step": 19
},
{
"epoch": 0.5,
"learning_rate": 2.6999999999999996e-05,
"loss": 13.9366,
"step": 20
},
{
"epoch": 0.53,
"learning_rate": 2.8499999999999998e-05,
"loss": 13.3282,
"step": 21
},
{
"epoch": 0.55,
"learning_rate": 2.9999999999999997e-05,
"loss": 12.1615,
"step": 22
},
{
"epoch": 0.57,
"learning_rate": 3.149999999999999e-05,
"loss": 11.4266,
"step": 23
},
{
"epoch": 0.6,
"learning_rate": 3.2999999999999996e-05,
"loss": 10.6892,
"step": 24
},
{
"epoch": 0.62,
"learning_rate": 3.45e-05,
"loss": 10.1604,
"step": 25
},
{
"epoch": 0.65,
"learning_rate": 3.45e-05,
"loss": 12.1081,
"step": 26
},
{
"epoch": 0.68,
"learning_rate": 3.5999999999999994e-05,
"loss": 11.3549,
"step": 27
},
{
"epoch": 0.7,
"learning_rate": 3.75e-05,
"loss": 10.1111,
"step": 28
},
{
"epoch": 0.72,
"learning_rate": 3.9e-05,
"loss": 9.0024,
"step": 29
},
{
"epoch": 0.75,
"learning_rate": 4.05e-05,
"loss": 8.6496,
"step": 30
},
{
"epoch": 0.78,
"learning_rate": 4.2e-05,
"loss": 8.0396,
"step": 31
},
{
"epoch": 0.8,
"learning_rate": 4.3499999999999993e-05,
"loss": 9.5749,
"step": 32
},
{
"epoch": 0.82,
"learning_rate": 4.4999999999999996e-05,
"loss": 8.4065,
"step": 33
},
{
"epoch": 0.85,
"learning_rate": 4.65e-05,
"loss": 7.891,
"step": 34
},
{
"epoch": 0.88,
"learning_rate": 4.7999999999999994e-05,
"loss": 7.0403,
"step": 35
},
{
"epoch": 0.9,
"learning_rate": 4.95e-05,
"loss": 7.1381,
"step": 36
},
{
"epoch": 0.93,
"learning_rate": 5.1e-05,
"loss": 6.6468,
"step": 37
},
{
"epoch": 0.95,
"learning_rate": 5.2499999999999995e-05,
"loss": 6.8392,
"step": 38
},
{
"epoch": 0.97,
"learning_rate": 5.399999999999999e-05,
"loss": 6.2868,
"step": 39
},
{
"epoch": 1.0,
"learning_rate": 5.5499999999999994e-05,
"loss": 5.9633,
"step": 40
},
{
"epoch": 1.02,
"learning_rate": 5.6999999999999996e-05,
"loss": 6.5684,
"step": 41
},
{
"epoch": 1.05,
"learning_rate": 5.85e-05,
"loss": 6.2034,
"step": 42
},
{
"epoch": 1.07,
"learning_rate": 5.9999999999999995e-05,
"loss": 5.5007,
"step": 43
},
{
"epoch": 1.1,
"learning_rate": 6.149999999999999e-05,
"loss": 5.3496,
"step": 44
},
{
"epoch": 1.12,
"learning_rate": 6.299999999999999e-05,
"loss": 5.4191,
"step": 45
},
{
"epoch": 1.15,
"learning_rate": 6.45e-05,
"loss": 5.135,
"step": 46
},
{
"epoch": 1.18,
"learning_rate": 6.599999999999999e-05,
"loss": 5.743,
"step": 47
},
{
"epoch": 1.2,
"learning_rate": 6.75e-05,
"loss": 5.3021,
"step": 48
},
{
"epoch": 1.23,
"learning_rate": 6.9e-05,
"loss": 5.2977,
"step": 49
},
{
"epoch": 1.25,
"learning_rate": 7.049999999999999e-05,
"loss": 4.901,
"step": 50
},
{
"epoch": 1.27,
"learning_rate": 7.199999999999999e-05,
"loss": 4.7971,
"step": 51
},
{
"epoch": 1.3,
"learning_rate": 7.35e-05,
"loss": 4.8289,
"step": 52
},
{
"epoch": 1.32,
"learning_rate": 7.5e-05,
"loss": 5.1127,
"step": 53
},
{
"epoch": 1.35,
"learning_rate": 7.649999999999999e-05,
"loss": 4.9781,
"step": 54
},
{
"epoch": 1.38,
"learning_rate": 7.8e-05,
"loss": 4.7937,
"step": 55
},
{
"epoch": 1.4,
"learning_rate": 7.95e-05,
"loss": 4.6379,
"step": 56
},
{
"epoch": 1.43,
"learning_rate": 8.1e-05,
"loss": 4.4712,
"step": 57
},
{
"epoch": 1.45,
"learning_rate": 8.25e-05,
"loss": 4.4839,
"step": 58
},
{
"epoch": 1.48,
"learning_rate": 8.4e-05,
"loss": 4.5183,
"step": 59
},
{
"epoch": 1.5,
"learning_rate": 8.549999999999999e-05,
"loss": 4.4879,
"step": 60
},
{
"epoch": 1.52,
"learning_rate": 8.699999999999999e-05,
"loss": 4.4971,
"step": 61
},
{
"epoch": 1.55,
"learning_rate": 8.849999999999998e-05,
"loss": 4.3831,
"step": 62
},
{
"epoch": 1.57,
"learning_rate": 8.999999999999999e-05,
"loss": 4.2891,
"step": 63
},
{
"epoch": 1.6,
"learning_rate": 9.149999999999999e-05,
"loss": 4.1934,
"step": 64
},
{
"epoch": 1.62,
"learning_rate": 9.3e-05,
"loss": 4.1251,
"step": 65
},
{
"epoch": 1.65,
"learning_rate": 9.449999999999999e-05,
"loss": 4.635,
"step": 66
},
{
"epoch": 1.68,
"learning_rate": 9.599999999999999e-05,
"loss": 4.165,
"step": 67
},
{
"epoch": 1.7,
"learning_rate": 9.75e-05,
"loss": 4.154,
"step": 68
},
{
"epoch": 1.73,
"learning_rate": 9.9e-05,
"loss": 4.0427,
"step": 69
},
{
"epoch": 1.75,
"learning_rate": 0.0001005,
"loss": 4.0911,
"step": 70
},
{
"epoch": 1.77,
"learning_rate": 0.000102,
"loss": 3.969,
"step": 71
},
{
"epoch": 1.8,
"learning_rate": 0.00010349999999999998,
"loss": 4.1288,
"step": 72
},
{
"epoch": 1.82,
"learning_rate": 0.00010499999999999999,
"loss": 4.091,
"step": 73
},
{
"epoch": 1.85,
"learning_rate": 0.00010649999999999999,
"loss": 3.9443,
"step": 74
},
{
"epoch": 1.88,
"learning_rate": 0.00010799999999999998,
"loss": 3.7792,
"step": 75
},
{
"epoch": 1.9,
"learning_rate": 0.00010949999999999999,
"loss": 3.8244,
"step": 76
},
{
"epoch": 1.93,
"learning_rate": 0.00011099999999999999,
"loss": 3.7284,
"step": 77
},
{
"epoch": 1.95,
"learning_rate": 0.0001125,
"loss": 3.8774,
"step": 78
},
{
"epoch": 1.98,
"learning_rate": 0.00011399999999999999,
"loss": 3.7435,
"step": 79
},
{
"epoch": 2.0,
"learning_rate": 0.00011549999999999999,
"loss": 3.6561,
"step": 80
},
{
"epoch": 2.02,
"learning_rate": 0.000117,
"loss": 3.9569,
"step": 81
},
{
"epoch": 2.05,
"learning_rate": 0.0001185,
"loss": 3.6893,
"step": 82
},
{
"epoch": 2.08,
"learning_rate": 0.00011999999999999999,
"loss": 3.6249,
"step": 83
},
{
"epoch": 2.1,
"learning_rate": 0.0001215,
"loss": 3.5364,
"step": 84
},
{
"epoch": 2.12,
"learning_rate": 0.00012299999999999998,
"loss": 3.5281,
"step": 85
},
{
"epoch": 2.15,
"learning_rate": 0.0001245,
"loss": 3.485,
"step": 86
},
{
"epoch": 2.17,
"learning_rate": 0.00012599999999999997,
"loss": 3.5419,
"step": 87
},
{
"epoch": 2.2,
"learning_rate": 0.00012749999999999998,
"loss": 3.4307,
"step": 88
},
{
"epoch": 2.23,
"learning_rate": 0.000129,
"loss": 3.4118,
"step": 89
},
{
"epoch": 2.25,
"learning_rate": 0.0001305,
"loss": 3.3756,
"step": 90
},
{
"epoch": 2.27,
"learning_rate": 0.00013199999999999998,
"loss": 3.3473,
"step": 91
},
{
"epoch": 2.3,
"learning_rate": 0.0001335,
"loss": 3.3208,
"step": 92
},
{
"epoch": 2.33,
"learning_rate": 0.000135,
"loss": 3.4177,
"step": 93
},
{
"epoch": 2.35,
"learning_rate": 0.00013649999999999998,
"loss": 3.3614,
"step": 94
},
{
"epoch": 2.38,
"learning_rate": 0.000138,
"loss": 3.3018,
"step": 95
},
{
"epoch": 2.4,
"learning_rate": 0.0001395,
"loss": 3.2517,
"step": 96
},
{
"epoch": 2.42,
"learning_rate": 0.00014099999999999998,
"loss": 3.2269,
"step": 97
},
{
"epoch": 2.45,
"learning_rate": 0.0001425,
"loss": 3.2305,
"step": 98
},
{
"epoch": 2.48,
"learning_rate": 0.00014399999999999998,
"loss": 3.2049,
"step": 99
},
{
"epoch": 2.5,
"learning_rate": 0.00014549999999999999,
"loss": 3.2521,
"step": 100
},
{
"epoch": 2.52,
"learning_rate": 0.000147,
"loss": 3.1998,
"step": 101
},
{
"epoch": 2.55,
"learning_rate": 0.00014849999999999998,
"loss": 3.1652,
"step": 102
},
{
"epoch": 2.58,
"learning_rate": 0.00015,
"loss": 3.1294,
"step": 103
},
{
"epoch": 2.6,
"learning_rate": 0.0001515,
"loss": 3.1336,
"step": 104
},
{
"epoch": 2.62,
"learning_rate": 0.00015299999999999998,
"loss": 3.1014,
"step": 105
},
{
"epoch": 2.65,
"learning_rate": 0.0001545,
"loss": 3.1631,
"step": 106
},
{
"epoch": 2.67,
"learning_rate": 0.000156,
"loss": 3.1484,
"step": 107
},
{
"epoch": 2.7,
"learning_rate": 0.00015749999999999998,
"loss": 3.0577,
"step": 108
},
{
"epoch": 2.73,
"learning_rate": 0.000159,
"loss": 3.0468,
"step": 109
},
{
"epoch": 2.75,
"learning_rate": 0.0001605,
"loss": 3.0804,
"step": 110
},
{
"epoch": 2.77,
"learning_rate": 0.000162,
"loss": 3.0399,
"step": 111
},
{
"epoch": 2.8,
"learning_rate": 0.0001635,
"loss": 3.0878,
"step": 112
},
{
"epoch": 2.83,
"learning_rate": 0.000165,
"loss": 3.0164,
"step": 113
},
{
"epoch": 2.85,
"learning_rate": 0.0001665,
"loss": 3.0297,
"step": 114
},
{
"epoch": 2.88,
"learning_rate": 0.000168,
"loss": 3.0057,
"step": 115
},
{
"epoch": 2.9,
"learning_rate": 0.00016949999999999997,
"loss": 2.9835,
"step": 116
},
{
"epoch": 2.92,
"learning_rate": 0.00017099999999999998,
"loss": 2.9727,
"step": 117
},
{
"epoch": 2.95,
"learning_rate": 0.00017249999999999996,
"loss": 3.007,
"step": 118
},
{
"epoch": 2.98,
"learning_rate": 0.00017399999999999997,
"loss": 2.9819,
"step": 119
},
{
"epoch": 3.0,
"learning_rate": 0.00017549999999999998,
"loss": 2.9878,
"step": 120
},
{
"epoch": 3.02,
"learning_rate": 0.00017699999999999997,
"loss": 3.0223,
"step": 121
},
{
"epoch": 3.05,
"learning_rate": 0.00017849999999999997,
"loss": 3.0296,
"step": 122
},
{
"epoch": 3.08,
"learning_rate": 0.00017999999999999998,
"loss": 2.9798,
"step": 123
},
{
"epoch": 3.1,
"learning_rate": 0.00018149999999999997,
"loss": 2.9757,
"step": 124
},
{
"epoch": 3.12,
"learning_rate": 0.00018299999999999998,
"loss": 2.9642,
"step": 125
},
{
"epoch": 3.15,
"learning_rate": 0.00018449999999999999,
"loss": 2.9522,
"step": 126
},
{
"epoch": 3.17,
"learning_rate": 0.000186,
"loss": 3.0148,
"step": 127
},
{
"epoch": 3.2,
"learning_rate": 0.00018749999999999998,
"loss": 2.9753,
"step": 128
},
{
"epoch": 3.23,
"learning_rate": 0.00018899999999999999,
"loss": 2.9676,
"step": 129
},
{
"epoch": 3.25,
"learning_rate": 0.0001905,
"loss": 2.9477,
"step": 130
},
{
"epoch": 3.27,
"learning_rate": 0.00019199999999999998,
"loss": 2.9437,
"step": 131
},
{
"epoch": 3.3,
"learning_rate": 0.0001935,
"loss": 2.9342,
"step": 132
},
{
"epoch": 3.33,
"learning_rate": 0.000195,
"loss": 2.9855,
"step": 133
},
{
"epoch": 3.35,
"learning_rate": 0.00019649999999999998,
"loss": 2.9948,
"step": 134
},
{
"epoch": 3.38,
"learning_rate": 0.000198,
"loss": 2.9306,
"step": 135
},
{
"epoch": 3.4,
"learning_rate": 0.0001995,
"loss": 2.9184,
"step": 136
},
{
"epoch": 3.42,
"learning_rate": 0.000201,
"loss": 2.9272,
"step": 137
},
{
"epoch": 3.45,
"learning_rate": 0.0002025,
"loss": 2.9779,
"step": 138
},
{
"epoch": 3.48,
"learning_rate": 0.000204,
"loss": 2.9396,
"step": 139
},
{
"epoch": 3.5,
"learning_rate": 0.0002055,
"loss": 2.9318,
"step": 140
},
{
"epoch": 3.52,
"learning_rate": 0.00020699999999999996,
"loss": 2.9382,
"step": 141
},
{
"epoch": 3.55,
"learning_rate": 0.00020849999999999997,
"loss": 2.9197,
"step": 142
},
{
"epoch": 3.58,
"learning_rate": 0.00020999999999999998,
"loss": 2.9319,
"step": 143
},
{
"epoch": 3.6,
"learning_rate": 0.00021149999999999996,
"loss": 2.9453,
"step": 144
},
{
"epoch": 3.62,
"learning_rate": 0.00021299999999999997,
"loss": 2.9234,
"step": 145
},
{
"epoch": 3.65,
"learning_rate": 0.00021449999999999998,
"loss": 2.93,
"step": 146
},
{
"epoch": 3.67,
"learning_rate": 0.00021599999999999996,
"loss": 2.9188,
"step": 147
},
{
"epoch": 3.7,
"learning_rate": 0.00021749999999999997,
"loss": 2.9095,
"step": 148
},
{
"epoch": 3.73,
"learning_rate": 0.00021899999999999998,
"loss": 2.9112,
"step": 149
},
{
"epoch": 3.75,
"learning_rate": 0.00022049999999999997,
"loss": 2.9158,
"step": 150
},
{
"epoch": 3.77,
"learning_rate": 0.00022199999999999998,
"loss": 2.9139,
"step": 151
},
{
"epoch": 3.8,
"learning_rate": 0.00022349999999999998,
"loss": 2.93,
"step": 152
},
{
"epoch": 3.83,
"learning_rate": 0.000225,
"loss": 2.9289,
"step": 153
},
{
"epoch": 3.85,
"learning_rate": 0.00022649999999999998,
"loss": 2.9068,
"step": 154
},
{
"epoch": 3.88,
"learning_rate": 0.00022799999999999999,
"loss": 2.9078,
"step": 155
},
{
"epoch": 3.9,
"learning_rate": 0.0002295,
"loss": 2.8934,
"step": 156
},
{
"epoch": 3.92,
"learning_rate": 0.00023099999999999998,
"loss": 2.9074,
"step": 157
},
{
"epoch": 3.95,
"learning_rate": 0.00023249999999999999,
"loss": 2.9112,
"step": 158
},
{
"epoch": 3.98,
"learning_rate": 0.000234,
"loss": 2.9088,
"step": 159
},
{
"epoch": 4.0,
"learning_rate": 0.00023549999999999998,
"loss": 2.9322,
"step": 160
},
{
"epoch": 4.03,
"learning_rate": 0.000237,
"loss": 2.9264,
"step": 161
},
{
"epoch": 4.05,
"learning_rate": 0.0002385,
"loss": 2.9052,
"step": 162
},
{
"epoch": 4.08,
"learning_rate": 0.00023999999999999998,
"loss": 2.8861,
"step": 163
},
{
"epoch": 4.1,
"learning_rate": 0.0002415,
"loss": 2.8886,
"step": 164
},
{
"epoch": 4.12,
"learning_rate": 0.000243,
"loss": 2.9083,
"step": 165
},
{
"epoch": 4.15,
"learning_rate": 0.0002445,
"loss": 2.9099,
"step": 166
},
{
"epoch": 4.17,
"learning_rate": 0.00024599999999999996,
"loss": 2.907,
"step": 167
},
{
"epoch": 4.2,
"learning_rate": 0.00024749999999999994,
"loss": 2.9047,
"step": 168
},
{
"epoch": 4.22,
"learning_rate": 0.000249,
"loss": 2.9063,
"step": 169
},
{
"epoch": 4.25,
"learning_rate": 0.00025049999999999996,
"loss": 2.8841,
"step": 170
},
{
"epoch": 4.28,
"learning_rate": 0.00025199999999999995,
"loss": 2.8924,
"step": 171
},
{
"epoch": 4.3,
"learning_rate": 0.0002535,
"loss": 2.8818,
"step": 172
},
{
"epoch": 4.33,
"learning_rate": 0.00025499999999999996,
"loss": 2.898,
"step": 173
},
{
"epoch": 4.35,
"learning_rate": 0.00025649999999999995,
"loss": 2.9063,
"step": 174
},
{
"epoch": 4.38,
"learning_rate": 0.000258,
"loss": 2.9082,
"step": 175
},
{
"epoch": 4.4,
"learning_rate": 0.00025949999999999997,
"loss": 2.8992,
"step": 176
},
{
"epoch": 4.42,
"learning_rate": 0.000261,
"loss": 2.9227,
"step": 177
},
{
"epoch": 4.45,
"learning_rate": 0.0002625,
"loss": 2.9063,
"step": 178
},
{
"epoch": 4.47,
"learning_rate": 0.00026399999999999997,
"loss": 2.8768,
"step": 179
},
{
"epoch": 4.5,
"learning_rate": 0.0002655,
"loss": 2.8997,
"step": 180
},
{
"epoch": 4.53,
"learning_rate": 0.000267,
"loss": 2.9128,
"step": 181
},
{
"epoch": 4.55,
"learning_rate": 0.00026849999999999997,
"loss": 2.871,
"step": 182
},
{
"epoch": 4.58,
"learning_rate": 0.00027,
"loss": 2.8746,
"step": 183
},
{
"epoch": 4.6,
"learning_rate": 0.0002715,
"loss": 2.8864,
"step": 184
},
{
"epoch": 4.62,
"learning_rate": 0.00027299999999999997,
"loss": 2.8741,
"step": 185
},
{
"epoch": 4.65,
"learning_rate": 0.0002745,
"loss": 2.9214,
"step": 186
},
{
"epoch": 4.67,
"learning_rate": 0.000276,
"loss": 2.9235,
"step": 187
},
{
"epoch": 4.7,
"learning_rate": 0.00027749999999999997,
"loss": 2.8978,
"step": 188
},
{
"epoch": 4.72,
"learning_rate": 0.000279,
"loss": 2.889,
"step": 189
},
{
"epoch": 4.75,
"learning_rate": 0.0002805,
"loss": 2.8871,
"step": 190
},
{
"epoch": 4.78,
"learning_rate": 0.00028199999999999997,
"loss": 2.8746,
"step": 191
},
{
"epoch": 4.8,
"learning_rate": 0.00028349999999999995,
"loss": 2.9017,
"step": 192
},
{
"epoch": 4.83,
"learning_rate": 0.000285,
"loss": 2.8951,
"step": 193
},
{
"epoch": 4.85,
"learning_rate": 0.00028649999999999997,
"loss": 2.8905,
"step": 194
},
{
"epoch": 4.88,
"learning_rate": 0.00028799999999999995,
"loss": 2.8908,
"step": 195
},
{
"epoch": 4.9,
"learning_rate": 0.0002895,
"loss": 2.8951,
"step": 196
},
{
"epoch": 4.92,
"learning_rate": 0.00029099999999999997,
"loss": 2.8982,
"step": 197
},
{
"epoch": 4.95,
"learning_rate": 0.00029249999999999995,
"loss": 2.8693,
"step": 198
},
{
"epoch": 4.97,
"learning_rate": 0.000294,
"loss": 2.8947,
"step": 199
},
{
"epoch": 5.0,
"learning_rate": 0.00029549999999999997,
"loss": 2.8769,
"step": 200
},
{
"epoch": 5.0,
"eval_cer": 0.9877725059047301,
"eval_loss": 2.887089252471924,
"eval_runtime": 7.9078,
"eval_samples_per_second": 49.824,
"eval_steps_per_second": 6.323,
"eval_wer": 1.0,
"step": 200
},
{
"epoch": 5.03,
"learning_rate": 0.00029699999999999996,
"loss": 2.9024,
"step": 201
},
{
"epoch": 5.05,
"learning_rate": 0.0002985,
"loss": 2.8893,
"step": 202
},
{
"epoch": 5.08,
"learning_rate": 0.0003,
"loss": 2.8794,
"step": 203
},
{
"epoch": 5.1,
"learning_rate": 0.00029969999999999997,
"loss": 2.8636,
"step": 204
},
{
"epoch": 5.12,
"learning_rate": 0.00029939999999999996,
"loss": 2.8694,
"step": 205
},
{
"epoch": 5.15,
"learning_rate": 0.00029909999999999995,
"loss": 2.856,
"step": 206
},
{
"epoch": 5.17,
"learning_rate": 0.0002988,
"loss": 2.8687,
"step": 207
},
{
"epoch": 5.2,
"learning_rate": 0.0002985,
"loss": 2.888,
"step": 208
},
{
"epoch": 5.22,
"learning_rate": 0.0002982,
"loss": 2.8859,
"step": 209
},
{
"epoch": 5.25,
"learning_rate": 0.0002979,
"loss": 2.8724,
"step": 210
},
{
"epoch": 5.28,
"learning_rate": 0.00029759999999999997,
"loss": 2.8656,
"step": 211
},
{
"epoch": 5.3,
"learning_rate": 0.00029729999999999996,
"loss": 2.8576,
"step": 212
},
{
"epoch": 5.33,
"learning_rate": 0.00029699999999999996,
"loss": 2.8593,
"step": 213
},
{
"epoch": 5.35,
"learning_rate": 0.00029669999999999995,
"loss": 2.8652,
"step": 214
},
{
"epoch": 5.38,
"learning_rate": 0.0002964,
"loss": 2.8831,
"step": 215
},
{
"epoch": 5.4,
"learning_rate": 0.0002961,
"loss": 2.8704,
"step": 216
},
{
"epoch": 5.42,
"learning_rate": 0.0002958,
"loss": 2.8628,
"step": 217
},
{
"epoch": 5.45,
"learning_rate": 0.00029549999999999997,
"loss": 2.878,
"step": 218
},
{
"epoch": 5.47,
"learning_rate": 0.00029519999999999997,
"loss": 2.857,
"step": 219
},
{
"epoch": 5.5,
"learning_rate": 0.00029489999999999996,
"loss": 2.8974,
"step": 220
},
{
"epoch": 5.53,
"learning_rate": 0.00029459999999999995,
"loss": 2.8828,
"step": 221
},
{
"epoch": 5.55,
"learning_rate": 0.00029429999999999994,
"loss": 2.8615,
"step": 222
},
{
"epoch": 5.58,
"learning_rate": 0.000294,
"loss": 2.8606,
"step": 223
},
{
"epoch": 5.6,
"learning_rate": 0.0002937,
"loss": 2.8739,
"step": 224
},
{
"epoch": 5.62,
"learning_rate": 0.0002934,
"loss": 2.8498,
"step": 225
},
{
"epoch": 5.65,
"learning_rate": 0.00029309999999999997,
"loss": 2.887,
"step": 226
},
{
"epoch": 5.67,
"learning_rate": 0.00029279999999999996,
"loss": 2.8769,
"step": 227
},
{
"epoch": 5.7,
"learning_rate": 0.00029249999999999995,
"loss": 2.8585,
"step": 228
},
{
"epoch": 5.72,
"learning_rate": 0.00029219999999999995,
"loss": 2.8652,
"step": 229
},
{
"epoch": 5.75,
"learning_rate": 0.0002919,
"loss": 2.8788,
"step": 230
},
{
"epoch": 5.78,
"learning_rate": 0.0002916,
"loss": 2.8695,
"step": 231
},
{
"epoch": 5.8,
"learning_rate": 0.0002913,
"loss": 2.8745,
"step": 232
},
{
"epoch": 5.83,
"learning_rate": 0.00029099999999999997,
"loss": 2.8641,
"step": 233
},
{
"epoch": 5.85,
"learning_rate": 0.00029069999999999996,
"loss": 2.8573,
"step": 234
},
{
"epoch": 5.88,
"learning_rate": 0.00029039999999999996,
"loss": 2.8544,
"step": 235
},
{
"epoch": 5.9,
"learning_rate": 0.00029009999999999995,
"loss": 2.8627,
"step": 236
},
{
"epoch": 5.92,
"learning_rate": 0.00028979999999999994,
"loss": 2.8689,
"step": 237
},
{
"epoch": 5.95,
"learning_rate": 0.0002895,
"loss": 2.8487,
"step": 238
},
{
"epoch": 5.97,
"learning_rate": 0.0002892,
"loss": 2.8712,
"step": 239
},
{
"epoch": 6.0,
"learning_rate": 0.0002889,
"loss": 2.8417,
"step": 240
},
{
"epoch": 6.03,
"learning_rate": 0.00028859999999999997,
"loss": 2.8779,
"step": 241
},
{
"epoch": 6.05,
"learning_rate": 0.00028829999999999996,
"loss": 2.856,
"step": 242
},
{
"epoch": 6.08,
"learning_rate": 0.00028799999999999995,
"loss": 2.8544,
"step": 243
},
{
"epoch": 6.1,
"learning_rate": 0.00028769999999999995,
"loss": 2.8747,
"step": 244
},
{
"epoch": 6.12,
"learning_rate": 0.00028739999999999994,
"loss": 2.8706,
"step": 245
},
{
"epoch": 6.15,
"learning_rate": 0.0002871,
"loss": 2.8637,
"step": 246
},
{
"epoch": 6.17,
"learning_rate": 0.0002868,
"loss": 2.8778,
"step": 247
},
{
"epoch": 6.2,
"learning_rate": 0.00028649999999999997,
"loss": 2.864,
"step": 248
},
{
"epoch": 6.22,
"learning_rate": 0.00028619999999999996,
"loss": 2.8515,
"step": 249
},
{
"epoch": 6.25,
"learning_rate": 0.00028589999999999996,
"loss": 2.8488,
"step": 250
},
{
"epoch": 6.28,
"learning_rate": 0.00028559999999999995,
"loss": 2.8588,
"step": 251
},
{
"epoch": 6.3,
"learning_rate": 0.00028529999999999994,
"loss": 2.8256,
"step": 252
},
{
"epoch": 6.33,
"learning_rate": 0.000285,
"loss": 2.8418,
"step": 253
},
{
"epoch": 6.35,
"learning_rate": 0.0002847,
"loss": 2.8455,
"step": 254
},
{
"epoch": 6.38,
"learning_rate": 0.0002844,
"loss": 2.8611,
"step": 255
},
{
"epoch": 6.4,
"learning_rate": 0.00028409999999999997,
"loss": 2.8389,
"step": 256
},
{
"epoch": 6.42,
"learning_rate": 0.00028379999999999996,
"loss": 2.8246,
"step": 257
},
{
"epoch": 6.45,
"learning_rate": 0.00028349999999999995,
"loss": 2.8091,
"step": 258
},
{
"epoch": 6.47,
"learning_rate": 0.00028319999999999994,
"loss": 2.8201,
"step": 259
},
{
"epoch": 6.5,
"learning_rate": 0.00028289999999999994,
"loss": 2.8465,
"step": 260
},
{
"epoch": 6.53,
"learning_rate": 0.0002826,
"loss": 2.8379,
"step": 261
},
{
"epoch": 6.55,
"learning_rate": 0.0002823,
"loss": 2.9108,
"step": 262
},
{
"epoch": 6.58,
"learning_rate": 0.00028199999999999997,
"loss": 2.8535,
"step": 263
},
{
"epoch": 6.6,
"learning_rate": 0.00028169999999999996,
"loss": 2.8134,
"step": 264
},
{
"epoch": 6.62,
"learning_rate": 0.00028139999999999996,
"loss": 2.7941,
"step": 265
},
{
"epoch": 6.65,
"learning_rate": 0.0002811,
"loss": 2.8202,
"step": 266
},
{
"epoch": 6.67,
"learning_rate": 0.0002808,
"loss": 2.7988,
"step": 267
},
{
"epoch": 6.7,
"learning_rate": 0.0002805,
"loss": 2.7547,
"step": 268
},
{
"epoch": 6.72,
"learning_rate": 0.0002802,
"loss": 2.7081,
"step": 269
},
{
"epoch": 6.75,
"learning_rate": 0.0002799,
"loss": 2.7188,
"step": 270
},
{
"epoch": 6.78,
"learning_rate": 0.00027959999999999997,
"loss": 2.6883,
"step": 271
},
{
"epoch": 6.8,
"learning_rate": 0.0002793,
"loss": 2.7156,
"step": 272
},
{
"epoch": 6.83,
"learning_rate": 0.000279,
"loss": 2.6894,
"step": 273
},
{
"epoch": 6.85,
"learning_rate": 0.0002787,
"loss": 2.6191,
"step": 274
},
{
"epoch": 6.88,
"learning_rate": 0.0002784,
"loss": 2.6832,
"step": 275
},
{
"epoch": 6.9,
"learning_rate": 0.0002781,
"loss": 2.5653,
"step": 276
},
{
"epoch": 6.92,
"learning_rate": 0.0002778,
"loss": 2.5862,
"step": 277
},
{
"epoch": 6.95,
"learning_rate": 0.00027749999999999997,
"loss": 2.4953,
"step": 278
},
{
"epoch": 6.97,
"learning_rate": 0.0002772,
"loss": 2.4946,
"step": 279
},
{
"epoch": 7.0,
"learning_rate": 0.0002769,
"loss": 2.4524,
"step": 280
},
{
"epoch": 7.03,
"learning_rate": 0.0002766,
"loss": 2.4326,
"step": 281
},
{
"epoch": 7.05,
"learning_rate": 0.0002763,
"loss": 2.3657,
"step": 282
},
{
"epoch": 7.08,
"learning_rate": 0.000276,
"loss": 2.2954,
"step": 283
},
{
"epoch": 7.1,
"learning_rate": 0.0002757,
"loss": 2.2269,
"step": 284
},
{
"epoch": 7.12,
"learning_rate": 0.00027539999999999997,
"loss": 2.2098,
"step": 285
},
{
"epoch": 7.15,
"learning_rate": 0.00027509999999999996,
"loss": 2.1506,
"step": 286
},
{
"epoch": 7.17,
"learning_rate": 0.0002748,
"loss": 2.2161,
"step": 287
},
{
"epoch": 7.2,
"learning_rate": 0.0002745,
"loss": 2.1576,
"step": 288
},
{
"epoch": 7.22,
"learning_rate": 0.0002742,
"loss": 2.0598,
"step": 289
},
{
"epoch": 7.25,
"learning_rate": 0.0002739,
"loss": 2.0789,
"step": 290
},
{
"epoch": 7.28,
"learning_rate": 0.0002736,
"loss": 1.9738,
"step": 291
},
{
"epoch": 7.3,
"learning_rate": 0.0002733,
"loss": 1.9215,
"step": 292
},
{
"epoch": 7.33,
"learning_rate": 0.00027299999999999997,
"loss": 1.8849,
"step": 293
},
{
"epoch": 7.35,
"learning_rate": 0.00027269999999999996,
"loss": 1.8478,
"step": 294
},
{
"epoch": 7.38,
"learning_rate": 0.0002724,
"loss": 1.795,
"step": 295
},
{
"epoch": 7.4,
"learning_rate": 0.0002721,
"loss": 1.7426,
"step": 296
},
{
"epoch": 7.42,
"learning_rate": 0.0002718,
"loss": 1.6475,
"step": 297
},
{
"epoch": 7.45,
"learning_rate": 0.0002715,
"loss": 1.5878,
"step": 298
},
{
"epoch": 7.47,
"learning_rate": 0.0002712,
"loss": 1.6167,
"step": 299
},
{
"epoch": 7.5,
"learning_rate": 0.00027089999999999997,
"loss": 1.6752,
"step": 300
},
{
"epoch": 7.53,
"learning_rate": 0.00027059999999999996,
"loss": 1.4976,
"step": 301
},
{
"epoch": 7.55,
"learning_rate": 0.00027029999999999996,
"loss": 1.4719,
"step": 302
},
{
"epoch": 7.58,
"learning_rate": 0.00027,
"loss": 1.4345,
"step": 303
},
{
"epoch": 7.6,
"learning_rate": 0.0002697,
"loss": 1.4122,
"step": 304
},
{
"epoch": 7.62,
"learning_rate": 0.0002694,
"loss": 1.2736,
"step": 305
},
{
"epoch": 7.65,
"learning_rate": 0.0002691,
"loss": 1.408,
"step": 306
},
{
"epoch": 7.67,
"learning_rate": 0.0002688,
"loss": 1.3904,
"step": 307
},
{
"epoch": 7.7,
"learning_rate": 0.00026849999999999997,
"loss": 1.3304,
"step": 308
},
{
"epoch": 7.72,
"learning_rate": 0.00026819999999999996,
"loss": 1.1912,
"step": 309
},
{
"epoch": 7.75,
"learning_rate": 0.0002679,
"loss": 1.1753,
"step": 310
},
{
"epoch": 7.78,
"learning_rate": 0.0002676,
"loss": 1.1732,
"step": 311
},
{
"epoch": 7.8,
"learning_rate": 0.0002673,
"loss": 1.2074,
"step": 312
},
{
"epoch": 7.83,
"learning_rate": 0.000267,
"loss": 1.1108,
"step": 313
},
{
"epoch": 7.85,
"learning_rate": 0.0002667,
"loss": 1.1075,
"step": 314
},
{
"epoch": 7.88,
"learning_rate": 0.00026639999999999997,
"loss": 1.0365,
"step": 315
},
{
"epoch": 7.9,
"learning_rate": 0.00026609999999999996,
"loss": 0.972,
"step": 316
},
{
"epoch": 7.92,
"learning_rate": 0.00026579999999999996,
"loss": 0.9697,
"step": 317
},
{
"epoch": 7.95,
"learning_rate": 0.0002655,
"loss": 0.9896,
"step": 318
},
{
"epoch": 7.97,
"learning_rate": 0.0002652,
"loss": 0.9626,
"step": 319
},
{
"epoch": 8.0,
"learning_rate": 0.0002649,
"loss": 0.985,
"step": 320
},
{
"epoch": 8.03,
"learning_rate": 0.0002646,
"loss": 0.9751,
"step": 321
},
{
"epoch": 8.05,
"learning_rate": 0.0002643,
"loss": 0.8559,
"step": 322
},
{
"epoch": 8.07,
"learning_rate": 0.00026399999999999997,
"loss": 0.8064,
"step": 323
},
{
"epoch": 8.1,
"learning_rate": 0.00026369999999999996,
"loss": 0.7447,
"step": 324
},
{
"epoch": 8.12,
"learning_rate": 0.00026339999999999995,
"loss": 0.7405,
"step": 325
},
{
"epoch": 8.15,
"learning_rate": 0.0002631,
"loss": 0.6602,
"step": 326
},
{
"epoch": 8.18,
"learning_rate": 0.0002628,
"loss": 0.7567,
"step": 327
},
{
"epoch": 8.2,
"learning_rate": 0.0002625,
"loss": 0.6912,
"step": 328
},
{
"epoch": 8.22,
"learning_rate": 0.0002622,
"loss": 0.7406,
"step": 329
},
{
"epoch": 8.25,
"learning_rate": 0.00026189999999999997,
"loss": 0.7396,
"step": 330
},
{
"epoch": 8.28,
"learning_rate": 0.00026159999999999996,
"loss": 0.6791,
"step": 331
},
{
"epoch": 8.3,
"learning_rate": 0.00026129999999999995,
"loss": 0.5948,
"step": 332
},
{
"epoch": 8.32,
"learning_rate": 0.000261,
"loss": 0.6729,
"step": 333
},
{
"epoch": 8.35,
"learning_rate": 0.0002607,
"loss": 0.6698,
"step": 334
},
{
"epoch": 8.38,
"learning_rate": 0.0002604,
"loss": 0.6305,
"step": 335
},
{
"epoch": 8.4,
"learning_rate": 0.0002601,
"loss": 0.6358,
"step": 336
},
{
"epoch": 8.43,
"learning_rate": 0.00025979999999999997,
"loss": 0.5804,
"step": 337
},
{
"epoch": 8.45,
"learning_rate": 0.00025949999999999997,
"loss": 0.5345,
"step": 338
},
{
"epoch": 8.47,
"learning_rate": 0.00025919999999999996,
"loss": 0.6047,
"step": 339
},
{
"epoch": 8.5,
"learning_rate": 0.00025889999999999995,
"loss": 0.656,
"step": 340
},
{
"epoch": 8.53,
"learning_rate": 0.0002586,
"loss": 0.557,
"step": 341
},
{
"epoch": 8.55,
"learning_rate": 0.0002583,
"loss": 0.5856,
"step": 342
},
{
"epoch": 8.57,
"learning_rate": 0.000258,
"loss": 0.5583,
"step": 343
},
{
"epoch": 8.6,
"learning_rate": 0.0002577,
"loss": 0.5335,
"step": 344
},
{
"epoch": 8.62,
"learning_rate": 0.00025739999999999997,
"loss": 0.4959,
"step": 345
},
{
"epoch": 8.65,
"learning_rate": 0.00025709999999999996,
"loss": 0.6238,
"step": 346
},
{
"epoch": 8.68,
"learning_rate": 0.00025679999999999995,
"loss": 0.6008,
"step": 347
},
{
"epoch": 8.7,
"learning_rate": 0.00025649999999999995,
"loss": 0.515,
"step": 348
},
{
"epoch": 8.72,
"learning_rate": 0.0002562,
"loss": 0.5541,
"step": 349
},
{
"epoch": 8.75,
"learning_rate": 0.0002559,
"loss": 0.5705,
"step": 350
},
{
"epoch": 8.78,
"learning_rate": 0.0002556,
"loss": 0.4859,
"step": 351
},
{
"epoch": 8.8,
"learning_rate": 0.00025529999999999997,
"loss": 0.5304,
"step": 352
},
{
"epoch": 8.82,
"learning_rate": 0.00025499999999999996,
"loss": 0.4834,
"step": 353
},
{
"epoch": 8.85,
"learning_rate": 0.00025469999999999996,
"loss": 0.5305,
"step": 354
},
{
"epoch": 8.88,
"learning_rate": 0.00025439999999999995,
"loss": 0.4857,
"step": 355
},
{
"epoch": 8.9,
"learning_rate": 0.0002541,
"loss": 0.4303,
"step": 356
},
{
"epoch": 8.93,
"learning_rate": 0.0002538,
"loss": 0.4325,
"step": 357
},
{
"epoch": 8.95,
"learning_rate": 0.0002535,
"loss": 0.545,
"step": 358
},
{
"epoch": 8.97,
"learning_rate": 0.0002532,
"loss": 0.4842,
"step": 359
},
{
"epoch": 9.0,
"learning_rate": 0.00025289999999999997,
"loss": 0.4255,
"step": 360
},
{
"epoch": 9.03,
"learning_rate": 0.00025259999999999996,
"loss": 0.4677,
"step": 361
},
{
"epoch": 9.05,
"learning_rate": 0.00025229999999999995,
"loss": 0.409,
"step": 362
},
{
"epoch": 9.07,
"learning_rate": 0.00025199999999999995,
"loss": 0.4293,
"step": 363
},
{
"epoch": 9.1,
"learning_rate": 0.0002517,
"loss": 0.3146,
"step": 364
},
{
"epoch": 9.12,
"learning_rate": 0.0002514,
"loss": 0.3329,
"step": 365
},
{
"epoch": 9.15,
"learning_rate": 0.0002511,
"loss": 0.3332,
"step": 366
},
{
"epoch": 9.18,
"learning_rate": 0.00025079999999999997,
"loss": 0.4191,
"step": 367
},
{
"epoch": 9.2,
"learning_rate": 0.00025049999999999996,
"loss": 0.4182,
"step": 368
},
{
"epoch": 9.22,
"learning_rate": 0.00025019999999999996,
"loss": 0.3925,
"step": 369
},
{
"epoch": 9.25,
"learning_rate": 0.00024989999999999995,
"loss": 0.3117,
"step": 370
},
{
"epoch": 9.28,
"learning_rate": 0.00024959999999999994,
"loss": 0.3054,
"step": 371
},
{
"epoch": 9.3,
"learning_rate": 0.0002493,
"loss": 0.3136,
"step": 372
},
{
"epoch": 9.32,
"learning_rate": 0.000249,
"loss": 0.3699,
"step": 373
},
{
"epoch": 9.35,
"learning_rate": 0.0002487,
"loss": 0.395,
"step": 374
},
{
"epoch": 9.38,
"learning_rate": 0.00024839999999999997,
"loss": 0.3453,
"step": 375
},
{
"epoch": 9.4,
"learning_rate": 0.00024809999999999996,
"loss": 0.3583,
"step": 376
},
{
"epoch": 9.43,
"learning_rate": 0.00024779999999999995,
"loss": 0.3036,
"step": 377
},
{
"epoch": 9.45,
"learning_rate": 0.00024749999999999994,
"loss": 0.286,
"step": 378
},
{
"epoch": 9.47,
"learning_rate": 0.0002472,
"loss": 0.2739,
"step": 379
},
{
"epoch": 9.5,
"learning_rate": 0.0002469,
"loss": 0.3236,
"step": 380
},
{
"epoch": 9.53,
"learning_rate": 0.0002466,
"loss": 0.3389,
"step": 381
},
{
"epoch": 9.55,
"learning_rate": 0.00024629999999999997,
"loss": 0.2755,
"step": 382
},
{
"epoch": 9.57,
"learning_rate": 0.00024599999999999996,
"loss": 0.3561,
"step": 383
},
{
"epoch": 9.6,
"learning_rate": 0.00024569999999999995,
"loss": 0.2761,
"step": 384
},
{
"epoch": 9.62,
"learning_rate": 0.00024539999999999995,
"loss": 0.3082,
"step": 385
},
{
"epoch": 9.65,
"learning_rate": 0.00024509999999999994,
"loss": 0.3891,
"step": 386
},
{
"epoch": 9.68,
"learning_rate": 0.0002448,
"loss": 0.3114,
"step": 387
},
{
"epoch": 9.7,
"learning_rate": 0.0002445,
"loss": 0.3427,
"step": 388
},
{
"epoch": 9.72,
"learning_rate": 0.00024419999999999997,
"loss": 0.3163,
"step": 389
},
{
"epoch": 9.75,
"learning_rate": 0.00024389999999999997,
"loss": 0.299,
"step": 390
},
{
"epoch": 9.78,
"learning_rate": 0.00024359999999999999,
"loss": 0.2514,
"step": 391
},
{
"epoch": 9.8,
"learning_rate": 0.0002433,
"loss": 0.3046,
"step": 392
},
{
"epoch": 9.82,
"learning_rate": 0.000243,
"loss": 0.313,
"step": 393
},
{
"epoch": 9.85,
"learning_rate": 0.0002427,
"loss": 0.3057,
"step": 394
},
{
"epoch": 9.88,
"learning_rate": 0.00024239999999999998,
"loss": 0.3002,
"step": 395
},
{
"epoch": 9.9,
"learning_rate": 0.0002421,
"loss": 0.2726,
"step": 396
},
{
"epoch": 9.93,
"learning_rate": 0.0002418,
"loss": 0.2354,
"step": 397
},
{
"epoch": 9.95,
"learning_rate": 0.0002415,
"loss": 0.3022,
"step": 398
},
{
"epoch": 9.97,
"learning_rate": 0.00024119999999999998,
"loss": 0.3126,
"step": 399
},
{
"epoch": 10.0,
"learning_rate": 0.0002409,
"loss": 0.2458,
"step": 400
},
{
"epoch": 10.0,
"eval_cer": 0.19513826474092344,
"eval_loss": 0.557033121585846,
"eval_runtime": 8.3651,
"eval_samples_per_second": 47.101,
"eval_steps_per_second": 5.977,
"eval_wer": 0.4898920997147464,
"step": 400
},
{
"epoch": 10.03,
"learning_rate": 0.0002406,
"loss": 0.3041,
"step": 401
},
{
"epoch": 10.05,
"learning_rate": 0.00024029999999999999,
"loss": 0.2244,
"step": 402
},
{
"epoch": 10.07,
"learning_rate": 0.00023999999999999998,
"loss": 0.2343,
"step": 403
},
{
"epoch": 10.1,
"learning_rate": 0.0002397,
"loss": 0.216,
"step": 404
},
{
"epoch": 10.12,
"learning_rate": 0.0002394,
"loss": 0.194,
"step": 405
},
{
"epoch": 10.15,
"learning_rate": 0.00023909999999999998,
"loss": 0.193,
"step": 406
},
{
"epoch": 10.18,
"learning_rate": 0.0002388,
"loss": 0.2353,
"step": 407
},
{
"epoch": 10.2,
"learning_rate": 0.0002385,
"loss": 0.2292,
"step": 408
},
{
"epoch": 10.22,
"learning_rate": 0.0002382,
"loss": 0.2217,
"step": 409
},
{
"epoch": 10.25,
"learning_rate": 0.00023789999999999998,
"loss": 0.2134,
"step": 410
},
{
"epoch": 10.28,
"learning_rate": 0.0002376,
"loss": 0.2036,
"step": 411
},
{
"epoch": 10.3,
"learning_rate": 0.0002373,
"loss": 0.2028,
"step": 412
},
{
"epoch": 10.32,
"learning_rate": 0.000237,
"loss": 0.2363,
"step": 413
},
{
"epoch": 10.35,
"learning_rate": 0.00023669999999999998,
"loss": 0.2264,
"step": 414
},
{
"epoch": 10.38,
"learning_rate": 0.0002364,
"loss": 0.2188,
"step": 415
},
{
"epoch": 10.4,
"learning_rate": 0.0002361,
"loss": 0.2323,
"step": 416
},
{
"epoch": 10.43,
"learning_rate": 0.00023579999999999999,
"loss": 0.2174,
"step": 417
},
{
"epoch": 10.45,
"learning_rate": 0.00023549999999999998,
"loss": 0.208,
"step": 418
},
{
"epoch": 10.47,
"learning_rate": 0.0002352,
"loss": 0.2065,
"step": 419
},
{
"epoch": 10.5,
"learning_rate": 0.0002349,
"loss": 0.2343,
"step": 420
},
{
"epoch": 10.53,
"learning_rate": 0.00023459999999999998,
"loss": 0.2301,
"step": 421
},
{
"epoch": 10.55,
"learning_rate": 0.00023429999999999998,
"loss": 0.2212,
"step": 422
},
{
"epoch": 10.57,
"learning_rate": 0.000234,
"loss": 0.2378,
"step": 423
},
{
"epoch": 10.6,
"learning_rate": 0.0002337,
"loss": 0.1698,
"step": 424
},
{
"epoch": 10.62,
"learning_rate": 0.00023339999999999998,
"loss": 0.1647,
"step": 425
},
{
"epoch": 10.65,
"learning_rate": 0.00023309999999999997,
"loss": 0.3171,
"step": 426
},
{
"epoch": 10.68,
"learning_rate": 0.0002328,
"loss": 0.2252,
"step": 427
},
{
"epoch": 10.7,
"learning_rate": 0.00023249999999999999,
"loss": 0.2558,
"step": 428
},
{
"epoch": 10.72,
"learning_rate": 0.00023219999999999998,
"loss": 0.2052,
"step": 429
},
{
"epoch": 10.75,
"learning_rate": 0.0002319,
"loss": 0.2069,
"step": 430
},
{
"epoch": 10.78,
"learning_rate": 0.0002316,
"loss": 0.1526,
"step": 431
},
{
"epoch": 10.8,
"learning_rate": 0.00023129999999999998,
"loss": 0.2558,
"step": 432
},
{
"epoch": 10.82,
"learning_rate": 0.00023099999999999998,
"loss": 0.2245,
"step": 433
},
{
"epoch": 10.85,
"learning_rate": 0.0002307,
"loss": 0.2275,
"step": 434
},
{
"epoch": 10.88,
"learning_rate": 0.0002304,
"loss": 0.1748,
"step": 435
},
{
"epoch": 10.9,
"learning_rate": 0.00023009999999999998,
"loss": 0.2545,
"step": 436
},
{
"epoch": 10.93,
"learning_rate": 0.00022979999999999997,
"loss": 0.1942,
"step": 437
},
{
"epoch": 10.95,
"learning_rate": 0.0002295,
"loss": 0.1936,
"step": 438
},
{
"epoch": 10.97,
"learning_rate": 0.0002292,
"loss": 0.1847,
"step": 439
},
{
"epoch": 11.0,
"learning_rate": 0.00022889999999999998,
"loss": 0.1901,
"step": 440
},
{
"epoch": 11.03,
"learning_rate": 0.00022859999999999997,
"loss": 0.2054,
"step": 441
},
{
"epoch": 11.05,
"learning_rate": 0.0002283,
"loss": 0.1832,
"step": 442
},
{
"epoch": 11.07,
"learning_rate": 0.00022799999999999999,
"loss": 0.1488,
"step": 443
},
{
"epoch": 11.1,
"learning_rate": 0.00022769999999999998,
"loss": 0.1574,
"step": 444
},
{
"epoch": 11.12,
"learning_rate": 0.00022739999999999997,
"loss": 0.1485,
"step": 445
},
{
"epoch": 11.15,
"learning_rate": 0.0002271,
"loss": 0.1413,
"step": 446
},
{
"epoch": 11.18,
"learning_rate": 0.00022679999999999998,
"loss": 0.1872,
"step": 447
},
{
"epoch": 11.2,
"learning_rate": 0.00022649999999999998,
"loss": 0.1605,
"step": 448
},
{
"epoch": 11.22,
"learning_rate": 0.00022619999999999997,
"loss": 0.1715,
"step": 449
},
{
"epoch": 11.25,
"learning_rate": 0.0002259,
"loss": 0.1592,
"step": 450
},
{
"epoch": 11.28,
"learning_rate": 0.00022559999999999998,
"loss": 0.1803,
"step": 451
},
{
"epoch": 11.3,
"learning_rate": 0.00022529999999999997,
"loss": 0.1331,
"step": 452
},
{
"epoch": 11.32,
"learning_rate": 0.000225,
"loss": 0.192,
"step": 453
},
{
"epoch": 11.35,
"learning_rate": 0.0002247,
"loss": 0.1627,
"step": 454
},
{
"epoch": 11.38,
"learning_rate": 0.00022439999999999998,
"loss": 0.1595,
"step": 455
},
{
"epoch": 11.4,
"learning_rate": 0.00022409999999999997,
"loss": 0.1496,
"step": 456
},
{
"epoch": 11.43,
"learning_rate": 0.0002238,
"loss": 0.1311,
"step": 457
},
{
"epoch": 11.45,
"learning_rate": 0.00022349999999999998,
"loss": 0.1518,
"step": 458
},
{
"epoch": 11.47,
"learning_rate": 0.00022319999999999998,
"loss": 0.1271,
"step": 459
},
{
"epoch": 11.5,
"learning_rate": 0.00022289999999999997,
"loss": 0.2051,
"step": 460
},
{
"epoch": 11.53,
"learning_rate": 0.0002226,
"loss": 0.1601,
"step": 461
},
{
"epoch": 11.55,
"learning_rate": 0.00022229999999999998,
"loss": 0.1355,
"step": 462
},
{
"epoch": 11.57,
"learning_rate": 0.00022199999999999998,
"loss": 0.1558,
"step": 463
},
{
"epoch": 11.6,
"learning_rate": 0.00022169999999999997,
"loss": 0.1173,
"step": 464
},
{
"epoch": 11.62,
"learning_rate": 0.0002214,
"loss": 0.1286,
"step": 465
},
{
"epoch": 11.65,
"learning_rate": 0.00022109999999999998,
"loss": 0.1992,
"step": 466
},
{
"epoch": 11.68,
"learning_rate": 0.00022079999999999997,
"loss": 0.1626,
"step": 467
},
{
"epoch": 11.7,
"learning_rate": 0.00022049999999999997,
"loss": 0.1253,
"step": 468
},
{
"epoch": 11.72,
"learning_rate": 0.00022019999999999999,
"loss": 0.1341,
"step": 469
},
{
"epoch": 11.75,
"learning_rate": 0.00021989999999999998,
"loss": 0.1199,
"step": 470
},
{
"epoch": 11.78,
"learning_rate": 0.00021959999999999997,
"loss": 0.1205,
"step": 471
},
{
"epoch": 11.8,
"learning_rate": 0.00021929999999999996,
"loss": 0.1673,
"step": 472
},
{
"epoch": 11.82,
"learning_rate": 0.00021899999999999998,
"loss": 0.171,
"step": 473
},
{
"epoch": 11.85,
"learning_rate": 0.00021869999999999998,
"loss": 0.1802,
"step": 474
},
{
"epoch": 11.88,
"learning_rate": 0.00021839999999999997,
"loss": 0.1373,
"step": 475
},
{
"epoch": 11.9,
"learning_rate": 0.00021809999999999996,
"loss": 0.1699,
"step": 476
},
{
"epoch": 11.93,
"learning_rate": 0.00021779999999999998,
"loss": 0.1432,
"step": 477
},
{
"epoch": 11.95,
"learning_rate": 0.00021749999999999997,
"loss": 0.154,
"step": 478
},
{
"epoch": 11.97,
"learning_rate": 0.00021719999999999997,
"loss": 0.1703,
"step": 479
},
{
"epoch": 12.0,
"learning_rate": 0.0002169,
"loss": 0.1231,
"step": 480
},
{
"epoch": 12.03,
"learning_rate": 0.00021659999999999998,
"loss": 0.1541,
"step": 481
},
{
"epoch": 12.05,
"learning_rate": 0.00021629999999999997,
"loss": 0.134,
"step": 482
},
{
"epoch": 12.07,
"learning_rate": 0.00021599999999999996,
"loss": 0.1277,
"step": 483
},
{
"epoch": 12.1,
"learning_rate": 0.00021569999999999998,
"loss": 0.1138,
"step": 484
},
{
"epoch": 12.12,
"learning_rate": 0.00021539999999999998,
"loss": 0.0952,
"step": 485
},
{
"epoch": 12.15,
"learning_rate": 0.00021509999999999997,
"loss": 0.0954,
"step": 486
},
{
"epoch": 12.18,
"learning_rate": 0.00021479999999999996,
"loss": 0.1573,
"step": 487
},
{
"epoch": 12.2,
"learning_rate": 0.00021449999999999998,
"loss": 0.1164,
"step": 488
},
{
"epoch": 12.22,
"learning_rate": 0.00021419999999999998,
"loss": 0.1351,
"step": 489
},
{
"epoch": 12.25,
"learning_rate": 0.00021389999999999997,
"loss": 0.121,
"step": 490
},
{
"epoch": 12.28,
"learning_rate": 0.00021359999999999996,
"loss": 0.0994,
"step": 491
},
{
"epoch": 12.3,
"learning_rate": 0.00021329999999999998,
"loss": 0.1037,
"step": 492
},
{
"epoch": 12.32,
"learning_rate": 0.00021299999999999997,
"loss": 0.1373,
"step": 493
},
{
"epoch": 12.35,
"learning_rate": 0.00021269999999999997,
"loss": 0.1446,
"step": 494
},
{
"epoch": 12.38,
"learning_rate": 0.00021239999999999996,
"loss": 0.1347,
"step": 495
},
{
"epoch": 12.4,
"learning_rate": 0.00021209999999999998,
"loss": 0.1213,
"step": 496
},
{
"epoch": 12.43,
"learning_rate": 0.00021179999999999997,
"loss": 0.1511,
"step": 497
},
{
"epoch": 12.45,
"learning_rate": 0.00021149999999999996,
"loss": 0.1132,
"step": 498
},
{
"epoch": 12.47,
"learning_rate": 0.00021119999999999996,
"loss": 0.096,
"step": 499
},
{
"epoch": 12.5,
"learning_rate": 0.00021089999999999998,
"loss": 0.1601,
"step": 500
},
{
"epoch": 12.53,
"learning_rate": 0.00021059999999999997,
"loss": 0.1313,
"step": 501
},
{
"epoch": 12.55,
"learning_rate": 0.00021029999999999996,
"loss": 0.1197,
"step": 502
},
{
"epoch": 12.57,
"learning_rate": 0.00020999999999999998,
"loss": 0.1094,
"step": 503
},
{
"epoch": 12.6,
"learning_rate": 0.00020969999999999997,
"loss": 0.1092,
"step": 504
},
{
"epoch": 12.62,
"learning_rate": 0.00020939999999999997,
"loss": 0.0848,
"step": 505
},
{
"epoch": 12.65,
"learning_rate": 0.00020909999999999996,
"loss": 0.1633,
"step": 506
},
{
"epoch": 12.68,
"learning_rate": 0.00020879999999999998,
"loss": 0.104,
"step": 507
},
{
"epoch": 12.7,
"learning_rate": 0.00020849999999999997,
"loss": 0.1511,
"step": 508
},
{
"epoch": 12.72,
"learning_rate": 0.00020819999999999996,
"loss": 0.1033,
"step": 509
},
{
"epoch": 12.75,
"learning_rate": 0.00020789999999999996,
"loss": 0.1007,
"step": 510
},
{
"epoch": 12.78,
"learning_rate": 0.00020759999999999998,
"loss": 0.1196,
"step": 511
},
{
"epoch": 12.8,
"learning_rate": 0.00020729999999999997,
"loss": 0.1394,
"step": 512
},
{
"epoch": 12.82,
"learning_rate": 0.00020699999999999996,
"loss": 0.1388,
"step": 513
},
{
"epoch": 12.85,
"learning_rate": 0.00020669999999999996,
"loss": 0.1097,
"step": 514
},
{
"epoch": 12.88,
"learning_rate": 0.00020639999999999998,
"loss": 0.1076,
"step": 515
},
{
"epoch": 12.9,
"learning_rate": 0.0002061,
"loss": 0.0821,
"step": 516
},
{
"epoch": 12.93,
"learning_rate": 0.0002058,
"loss": 0.1071,
"step": 517
},
{
"epoch": 12.95,
"learning_rate": 0.0002055,
"loss": 0.1106,
"step": 518
},
{
"epoch": 12.97,
"learning_rate": 0.0002052,
"loss": 0.1335,
"step": 519
},
{
"epoch": 13.0,
"learning_rate": 0.0002049,
"loss": 0.1112,
"step": 520
},
{
"epoch": 13.03,
"learning_rate": 0.00020459999999999999,
"loss": 0.1057,
"step": 521
},
{
"epoch": 13.05,
"learning_rate": 0.0002043,
"loss": 0.1022,
"step": 522
},
{
"epoch": 13.07,
"learning_rate": 0.000204,
"loss": 0.1005,
"step": 523
},
{
"epoch": 13.1,
"learning_rate": 0.0002037,
"loss": 0.0933,
"step": 524
},
{
"epoch": 13.12,
"learning_rate": 0.00020339999999999998,
"loss": 0.1216,
"step": 525
},
{
"epoch": 13.15,
"learning_rate": 0.0002031,
"loss": 0.0897,
"step": 526
},
{
"epoch": 13.18,
"learning_rate": 0.0002028,
"loss": 0.1077,
"step": 527
},
{
"epoch": 13.2,
"learning_rate": 0.0002025,
"loss": 0.1295,
"step": 528
},
{
"epoch": 13.22,
"learning_rate": 0.0002022,
"loss": 0.1115,
"step": 529
},
{
"epoch": 13.25,
"learning_rate": 0.0002019,
"loss": 0.0805,
"step": 530
},
{
"epoch": 13.28,
"learning_rate": 0.0002016,
"loss": 0.0825,
"step": 531
},
{
"epoch": 13.3,
"learning_rate": 0.0002013,
"loss": 0.077,
"step": 532
},
{
"epoch": 13.32,
"learning_rate": 0.000201,
"loss": 0.1096,
"step": 533
},
{
"epoch": 13.35,
"learning_rate": 0.0002007,
"loss": 0.1062,
"step": 534
},
{
"epoch": 13.38,
"learning_rate": 0.0002004,
"loss": 0.118,
"step": 535
},
{
"epoch": 13.4,
"learning_rate": 0.00020009999999999998,
"loss": 0.0983,
"step": 536
},
{
"epoch": 13.43,
"learning_rate": 0.0001998,
"loss": 0.0855,
"step": 537
},
{
"epoch": 13.45,
"learning_rate": 0.0001995,
"loss": 0.0937,
"step": 538
},
{
"epoch": 13.47,
"learning_rate": 0.0001992,
"loss": 0.106,
"step": 539
},
{
"epoch": 13.5,
"learning_rate": 0.00019889999999999998,
"loss": 0.1077,
"step": 540
},
{
"epoch": 13.53,
"learning_rate": 0.0001986,
"loss": 0.0825,
"step": 541
},
{
"epoch": 13.55,
"learning_rate": 0.0001983,
"loss": 0.1028,
"step": 542
},
{
"epoch": 13.57,
"learning_rate": 0.000198,
"loss": 0.0723,
"step": 543
},
{
"epoch": 13.6,
"learning_rate": 0.00019769999999999998,
"loss": 0.0702,
"step": 544
},
{
"epoch": 13.62,
"learning_rate": 0.0001974,
"loss": 0.0939,
"step": 545
},
{
"epoch": 13.65,
"learning_rate": 0.0001971,
"loss": 0.1375,
"step": 546
},
{
"epoch": 13.68,
"learning_rate": 0.00019679999999999999,
"loss": 0.1154,
"step": 547
},
{
"epoch": 13.7,
"learning_rate": 0.00019649999999999998,
"loss": 0.067,
"step": 548
},
{
"epoch": 13.72,
"learning_rate": 0.0001962,
"loss": 0.0911,
"step": 549
},
{
"epoch": 13.75,
"learning_rate": 0.0001959,
"loss": 0.0685,
"step": 550
},
{
"epoch": 13.78,
"learning_rate": 0.00019559999999999998,
"loss": 0.0861,
"step": 551
},
{
"epoch": 13.8,
"learning_rate": 0.00019529999999999998,
"loss": 0.1077,
"step": 552
},
{
"epoch": 13.82,
"learning_rate": 0.000195,
"loss": 0.102,
"step": 553
},
{
"epoch": 13.85,
"learning_rate": 0.0001947,
"loss": 0.0851,
"step": 554
},
{
"epoch": 13.88,
"learning_rate": 0.00019439999999999998,
"loss": 0.0936,
"step": 555
},
{
"epoch": 13.9,
"learning_rate": 0.0001941,
"loss": 0.0976,
"step": 556
},
{
"epoch": 13.93,
"learning_rate": 0.0001938,
"loss": 0.0886,
"step": 557
},
{
"epoch": 13.95,
"learning_rate": 0.0001935,
"loss": 0.0929,
"step": 558
},
{
"epoch": 13.97,
"learning_rate": 0.00019319999999999998,
"loss": 0.1077,
"step": 559
},
{
"epoch": 14.0,
"learning_rate": 0.0001929,
"loss": 0.0969,
"step": 560
},
{
"epoch": 14.03,
"learning_rate": 0.0001926,
"loss": 0.1275,
"step": 561
},
{
"epoch": 14.05,
"learning_rate": 0.00019229999999999999,
"loss": 0.0881,
"step": 562
},
{
"epoch": 14.07,
"learning_rate": 0.00019199999999999998,
"loss": 0.0876,
"step": 563
},
{
"epoch": 14.1,
"learning_rate": 0.0001917,
"loss": 0.0816,
"step": 564
},
{
"epoch": 14.12,
"learning_rate": 0.0001914,
"loss": 0.0606,
"step": 565
},
{
"epoch": 14.15,
"learning_rate": 0.00019109999999999998,
"loss": 0.0819,
"step": 566
},
{
"epoch": 14.18,
"learning_rate": 0.00019079999999999998,
"loss": 0.0884,
"step": 567
},
{
"epoch": 14.2,
"learning_rate": 0.0001905,
"loss": 0.0923,
"step": 568
},
{
"epoch": 14.22,
"learning_rate": 0.0001902,
"loss": 0.0595,
"step": 569
},
{
"epoch": 14.25,
"learning_rate": 0.00018989999999999998,
"loss": 0.0796,
"step": 570
},
{
"epoch": 14.28,
"learning_rate": 0.00018959999999999997,
"loss": 0.0635,
"step": 571
},
{
"epoch": 14.3,
"learning_rate": 0.0001893,
"loss": 0.0728,
"step": 572
},
{
"epoch": 14.32,
"learning_rate": 0.00018899999999999999,
"loss": 0.1051,
"step": 573
},
{
"epoch": 14.35,
"learning_rate": 0.00018869999999999998,
"loss": 0.0853,
"step": 574
},
{
"epoch": 14.38,
"learning_rate": 0.00018839999999999997,
"loss": 0.0846,
"step": 575
},
{
"epoch": 14.4,
"learning_rate": 0.0001881,
"loss": 0.0869,
"step": 576
},
{
"epoch": 14.43,
"learning_rate": 0.00018779999999999998,
"loss": 0.064,
"step": 577
},
{
"epoch": 14.45,
"learning_rate": 0.00018749999999999998,
"loss": 0.0668,
"step": 578
},
{
"epoch": 14.47,
"learning_rate": 0.0001872,
"loss": 0.0847,
"step": 579
},
{
"epoch": 14.5,
"learning_rate": 0.0001869,
"loss": 0.1166,
"step": 580
},
{
"epoch": 14.53,
"learning_rate": 0.00018659999999999998,
"loss": 0.0903,
"step": 581
},
{
"epoch": 14.55,
"learning_rate": 0.00018629999999999997,
"loss": 0.1085,
"step": 582
},
{
"epoch": 14.57,
"learning_rate": 0.000186,
"loss": 0.1275,
"step": 583
},
{
"epoch": 14.6,
"learning_rate": 0.0001857,
"loss": 0.0655,
"step": 584
},
{
"epoch": 14.62,
"learning_rate": 0.00018539999999999998,
"loss": 0.0636,
"step": 585
},
{
"epoch": 14.65,
"learning_rate": 0.00018509999999999997,
"loss": 0.1008,
"step": 586
},
{
"epoch": 14.68,
"learning_rate": 0.0001848,
"loss": 0.1095,
"step": 587
},
{
"epoch": 14.7,
"learning_rate": 0.00018449999999999999,
"loss": 0.0729,
"step": 588
},
{
"epoch": 14.72,
"learning_rate": 0.00018419999999999998,
"loss": 0.0819,
"step": 589
},
{
"epoch": 14.75,
"learning_rate": 0.00018389999999999997,
"loss": 0.0807,
"step": 590
},
{
"epoch": 14.78,
"learning_rate": 0.0001836,
"loss": 0.0503,
"step": 591
},
{
"epoch": 14.8,
"learning_rate": 0.00018329999999999998,
"loss": 0.1053,
"step": 592
},
{
"epoch": 14.82,
"learning_rate": 0.00018299999999999998,
"loss": 0.0826,
"step": 593
},
{
"epoch": 14.85,
"learning_rate": 0.00018269999999999997,
"loss": 0.0898,
"step": 594
},
{
"epoch": 14.88,
"learning_rate": 0.0001824,
"loss": 0.0703,
"step": 595
},
{
"epoch": 14.9,
"learning_rate": 0.00018209999999999998,
"loss": 0.0703,
"step": 596
},
{
"epoch": 14.93,
"learning_rate": 0.00018179999999999997,
"loss": 0.0759,
"step": 597
},
{
"epoch": 14.95,
"learning_rate": 0.00018149999999999997,
"loss": 0.0918,
"step": 598
},
{
"epoch": 14.97,
"learning_rate": 0.00018119999999999999,
"loss": 0.0818,
"step": 599
},
{
"epoch": 15.0,
"learning_rate": 0.00018089999999999998,
"loss": 0.0762,
"step": 600
},
{
"epoch": 15.0,
"eval_cer": 0.15617749723052485,
"eval_loss": 0.5212965607643127,
"eval_runtime": 8.4311,
"eval_samples_per_second": 46.732,
"eval_steps_per_second": 5.93,
"eval_wer": 0.3726900657323577,
"step": 600
},
{
"epoch": 15.03,
"learning_rate": 0.00018059999999999997,
"loss": 0.1135,
"step": 601
},
{
"epoch": 15.05,
"learning_rate": 0.00018029999999999996,
"loss": 0.0773,
"step": 602
},
{
"epoch": 15.07,
"learning_rate": 0.00017999999999999998,
"loss": 0.0751,
"step": 603
},
{
"epoch": 15.1,
"learning_rate": 0.00017969999999999998,
"loss": 0.064,
"step": 604
},
{
"epoch": 15.12,
"learning_rate": 0.00017939999999999997,
"loss": 0.0756,
"step": 605
},
{
"epoch": 15.15,
"learning_rate": 0.0001791,
"loss": 0.0821,
"step": 606
},
{
"epoch": 15.18,
"learning_rate": 0.00017879999999999998,
"loss": 0.0922,
"step": 607
},
{
"epoch": 15.2,
"learning_rate": 0.00017849999999999997,
"loss": 0.0719,
"step": 608
},
{
"epoch": 15.22,
"learning_rate": 0.00017819999999999997,
"loss": 0.0729,
"step": 609
},
{
"epoch": 15.25,
"learning_rate": 0.0001779,
"loss": 0.1057,
"step": 610
},
{
"epoch": 15.28,
"learning_rate": 0.00017759999999999998,
"loss": 0.1273,
"step": 611
},
{
"epoch": 15.3,
"learning_rate": 0.00017729999999999997,
"loss": 0.0513,
"step": 612
},
{
"epoch": 15.32,
"learning_rate": 0.00017699999999999997,
"loss": 0.0757,
"step": 613
},
{
"epoch": 15.35,
"learning_rate": 0.00017669999999999999,
"loss": 0.0734,
"step": 614
},
{
"epoch": 15.38,
"learning_rate": 0.00017639999999999998,
"loss": 0.0682,
"step": 615
},
{
"epoch": 15.4,
"learning_rate": 0.00017609999999999997,
"loss": 0.0755,
"step": 616
},
{
"epoch": 15.43,
"learning_rate": 0.00017579999999999996,
"loss": 0.0568,
"step": 617
},
{
"epoch": 15.45,
"learning_rate": 0.00017549999999999998,
"loss": 0.0719,
"step": 618
},
{
"epoch": 15.47,
"learning_rate": 0.00017519999999999998,
"loss": 0.068,
"step": 619
},
{
"epoch": 15.5,
"learning_rate": 0.00017489999999999997,
"loss": 0.0963,
"step": 620
},
{
"epoch": 15.53,
"learning_rate": 0.00017459999999999996,
"loss": 0.0916,
"step": 621
},
{
"epoch": 15.55,
"learning_rate": 0.00017429999999999998,
"loss": 0.0672,
"step": 622
},
{
"epoch": 15.57,
"learning_rate": 0.00017399999999999997,
"loss": 0.0784,
"step": 623
},
{
"epoch": 15.6,
"learning_rate": 0.00017369999999999997,
"loss": 0.0672,
"step": 624
},
{
"epoch": 15.62,
"learning_rate": 0.00017339999999999996,
"loss": 0.0729,
"step": 625
},
{
"epoch": 15.65,
"learning_rate": 0.00017309999999999998,
"loss": 0.0833,
"step": 626
},
{
"epoch": 15.68,
"learning_rate": 0.00017279999999999997,
"loss": 0.0799,
"step": 627
},
{
"epoch": 15.7,
"learning_rate": 0.00017249999999999996,
"loss": 0.0448,
"step": 628
},
{
"epoch": 15.72,
"learning_rate": 0.00017219999999999998,
"loss": 0.0702,
"step": 629
},
{
"epoch": 15.75,
"learning_rate": 0.00017189999999999998,
"loss": 0.0668,
"step": 630
},
{
"epoch": 15.78,
"learning_rate": 0.00017159999999999997,
"loss": 0.0719,
"step": 631
},
{
"epoch": 15.8,
"learning_rate": 0.00017129999999999996,
"loss": 0.0799,
"step": 632
},
{
"epoch": 15.82,
"learning_rate": 0.00017099999999999998,
"loss": 0.0898,
"step": 633
},
{
"epoch": 15.85,
"learning_rate": 0.00017069999999999998,
"loss": 0.0656,
"step": 634
},
{
"epoch": 15.88,
"learning_rate": 0.00017039999999999997,
"loss": 0.063,
"step": 635
},
{
"epoch": 15.9,
"learning_rate": 0.00017009999999999996,
"loss": 0.0493,
"step": 636
},
{
"epoch": 15.93,
"learning_rate": 0.00016979999999999998,
"loss": 0.0872,
"step": 637
},
{
"epoch": 15.95,
"learning_rate": 0.00016949999999999997,
"loss": 0.0663,
"step": 638
},
{
"epoch": 15.97,
"learning_rate": 0.00016919999999999997,
"loss": 0.0779,
"step": 639
},
{
"epoch": 16.0,
"learning_rate": 0.00016889999999999996,
"loss": 0.059,
"step": 640
},
{
"epoch": 16.02,
"learning_rate": 0.0001686,
"loss": 0.0903,
"step": 641
},
{
"epoch": 16.05,
"learning_rate": 0.0001683,
"loss": 0.0713,
"step": 642
},
{
"epoch": 16.07,
"learning_rate": 0.000168,
"loss": 0.0696,
"step": 643
},
{
"epoch": 16.1,
"learning_rate": 0.0001677,
"loss": 0.0527,
"step": 644
},
{
"epoch": 16.12,
"learning_rate": 0.0001674,
"loss": 0.0756,
"step": 645
},
{
"epoch": 16.15,
"learning_rate": 0.0001671,
"loss": 0.051,
"step": 646
},
{
"epoch": 16.18,
"learning_rate": 0.0001668,
"loss": 0.0809,
"step": 647
},
{
"epoch": 16.2,
"learning_rate": 0.0001665,
"loss": 0.0698,
"step": 648
},
{
"epoch": 16.23,
"learning_rate": 0.0001662,
"loss": 0.0686,
"step": 649
},
{
"epoch": 16.25,
"learning_rate": 0.0001659,
"loss": 0.0572,
"step": 650
},
{
"epoch": 16.27,
"learning_rate": 0.0001656,
"loss": 0.0551,
"step": 651
},
{
"epoch": 16.3,
"learning_rate": 0.0001653,
"loss": 0.065,
"step": 652
},
{
"epoch": 16.32,
"learning_rate": 0.000165,
"loss": 0.0693,
"step": 653
},
{
"epoch": 16.35,
"learning_rate": 0.0001647,
"loss": 0.0908,
"step": 654
},
{
"epoch": 16.38,
"learning_rate": 0.0001644,
"loss": 0.098,
"step": 655
},
{
"epoch": 16.4,
"learning_rate": 0.0001641,
"loss": 0.0605,
"step": 656
},
{
"epoch": 16.43,
"learning_rate": 0.0001638,
"loss": 0.0866,
"step": 657
},
{
"epoch": 16.45,
"learning_rate": 0.0001635,
"loss": 0.0523,
"step": 658
},
{
"epoch": 16.48,
"learning_rate": 0.0001632,
"loss": 0.077,
"step": 659
},
{
"epoch": 16.5,
"learning_rate": 0.0001629,
"loss": 0.0685,
"step": 660
},
{
"epoch": 16.52,
"learning_rate": 0.0001626,
"loss": 0.0561,
"step": 661
},
{
"epoch": 16.55,
"learning_rate": 0.0001623,
"loss": 0.0857,
"step": 662
},
{
"epoch": 16.57,
"learning_rate": 0.000162,
"loss": 0.0643,
"step": 663
},
{
"epoch": 16.6,
"learning_rate": 0.0001617,
"loss": 0.0518,
"step": 664
},
{
"epoch": 16.62,
"learning_rate": 0.0001614,
"loss": 0.0349,
"step": 665
},
{
"epoch": 16.65,
"learning_rate": 0.00016109999999999999,
"loss": 0.0626,
"step": 666
},
{
"epoch": 16.68,
"learning_rate": 0.0001608,
"loss": 0.0781,
"step": 667
},
{
"epoch": 16.7,
"learning_rate": 0.0001605,
"loss": 0.0667,
"step": 668
},
{
"epoch": 16.73,
"learning_rate": 0.0001602,
"loss": 0.0616,
"step": 669
},
{
"epoch": 16.75,
"learning_rate": 0.00015989999999999998,
"loss": 0.0693,
"step": 670
},
{
"epoch": 16.77,
"learning_rate": 0.0001596,
"loss": 0.0614,
"step": 671
},
{
"epoch": 16.8,
"learning_rate": 0.0001593,
"loss": 0.0895,
"step": 672
},
{
"epoch": 16.82,
"learning_rate": 0.000159,
"loss": 0.0742,
"step": 673
},
{
"epoch": 16.85,
"learning_rate": 0.00015869999999999998,
"loss": 0.0619,
"step": 674
},
{
"epoch": 16.88,
"learning_rate": 0.0001584,
"loss": 0.0557,
"step": 675
},
{
"epoch": 16.9,
"learning_rate": 0.0001581,
"loss": 0.0538,
"step": 676
},
{
"epoch": 16.93,
"learning_rate": 0.0001578,
"loss": 0.0737,
"step": 677
},
{
"epoch": 16.95,
"learning_rate": 0.00015749999999999998,
"loss": 0.072,
"step": 678
},
{
"epoch": 16.98,
"learning_rate": 0.0001572,
"loss": 0.08,
"step": 679
},
{
"epoch": 17.0,
"learning_rate": 0.0001569,
"loss": 0.0508,
"step": 680
},
{
"epoch": 17.02,
"learning_rate": 0.00015659999999999998,
"loss": 0.0787,
"step": 681
},
{
"epoch": 17.05,
"learning_rate": 0.0001563,
"loss": 0.0546,
"step": 682
},
{
"epoch": 17.07,
"learning_rate": 0.000156,
"loss": 0.0689,
"step": 683
},
{
"epoch": 17.1,
"learning_rate": 0.0001557,
"loss": 0.0617,
"step": 684
},
{
"epoch": 17.12,
"learning_rate": 0.00015539999999999998,
"loss": 0.0422,
"step": 685
},
{
"epoch": 17.15,
"learning_rate": 0.0001551,
"loss": 0.067,
"step": 686
},
{
"epoch": 17.18,
"learning_rate": 0.0001548,
"loss": 0.071,
"step": 687
},
{
"epoch": 17.2,
"learning_rate": 0.0001545,
"loss": 0.0647,
"step": 688
},
{
"epoch": 17.23,
"learning_rate": 0.00015419999999999998,
"loss": 0.0582,
"step": 689
},
{
"epoch": 17.25,
"learning_rate": 0.0001539,
"loss": 0.0481,
"step": 690
},
{
"epoch": 17.27,
"learning_rate": 0.0001536,
"loss": 0.0874,
"step": 691
},
{
"epoch": 17.3,
"learning_rate": 0.00015329999999999999,
"loss": 0.0575,
"step": 692
},
{
"epoch": 17.32,
"learning_rate": 0.00015299999999999998,
"loss": 0.0634,
"step": 693
},
{
"epoch": 17.35,
"learning_rate": 0.0001527,
"loss": 0.0515,
"step": 694
},
{
"epoch": 17.38,
"learning_rate": 0.0001524,
"loss": 0.0585,
"step": 695
},
{
"epoch": 17.4,
"learning_rate": 0.00015209999999999998,
"loss": 0.0551,
"step": 696
},
{
"epoch": 17.43,
"learning_rate": 0.00015179999999999998,
"loss": 0.0392,
"step": 697
},
{
"epoch": 17.45,
"learning_rate": 0.0001515,
"loss": 0.0543,
"step": 698
},
{
"epoch": 17.48,
"learning_rate": 0.0001512,
"loss": 0.0577,
"step": 699
},
{
"epoch": 17.5,
"learning_rate": 0.00015089999999999998,
"loss": 0.0713,
"step": 700
},
{
"epoch": 17.52,
"learning_rate": 0.00015059999999999997,
"loss": 0.0585,
"step": 701
},
{
"epoch": 17.55,
"learning_rate": 0.0001503,
"loss": 0.0689,
"step": 702
},
{
"epoch": 17.57,
"learning_rate": 0.00015,
"loss": 0.0691,
"step": 703
},
{
"epoch": 17.6,
"learning_rate": 0.00014969999999999998,
"loss": 0.0418,
"step": 704
},
{
"epoch": 17.62,
"learning_rate": 0.0001494,
"loss": 0.0588,
"step": 705
},
{
"epoch": 17.65,
"learning_rate": 0.0001491,
"loss": 0.0716,
"step": 706
},
{
"epoch": 17.68,
"learning_rate": 0.00014879999999999998,
"loss": 0.0642,
"step": 707
},
{
"epoch": 17.7,
"learning_rate": 0.00014849999999999998,
"loss": 0.0507,
"step": 708
},
{
"epoch": 17.73,
"learning_rate": 0.0001482,
"loss": 0.0484,
"step": 709
},
{
"epoch": 17.75,
"learning_rate": 0.0001479,
"loss": 0.0524,
"step": 710
},
{
"epoch": 17.77,
"learning_rate": 0.00014759999999999998,
"loss": 0.0416,
"step": 711
},
{
"epoch": 17.8,
"learning_rate": 0.00014729999999999998,
"loss": 0.0626,
"step": 712
},
{
"epoch": 17.82,
"learning_rate": 0.000147,
"loss": 0.0815,
"step": 713
},
{
"epoch": 17.85,
"learning_rate": 0.0001467,
"loss": 0.0653,
"step": 714
},
{
"epoch": 17.88,
"learning_rate": 0.00014639999999999998,
"loss": 0.0433,
"step": 715
},
{
"epoch": 17.9,
"learning_rate": 0.00014609999999999997,
"loss": 0.0512,
"step": 716
},
{
"epoch": 17.93,
"learning_rate": 0.0001458,
"loss": 0.051,
"step": 717
},
{
"epoch": 17.95,
"learning_rate": 0.00014549999999999999,
"loss": 0.0628,
"step": 718
},
{
"epoch": 17.98,
"learning_rate": 0.00014519999999999998,
"loss": 0.0643,
"step": 719
},
{
"epoch": 18.0,
"learning_rate": 0.00014489999999999997,
"loss": 0.0533,
"step": 720
},
{
"epoch": 18.02,
"learning_rate": 0.0001446,
"loss": 0.0536,
"step": 721
},
{
"epoch": 18.05,
"learning_rate": 0.00014429999999999998,
"loss": 0.0526,
"step": 722
},
{
"epoch": 18.07,
"learning_rate": 0.00014399999999999998,
"loss": 0.0506,
"step": 723
},
{
"epoch": 18.1,
"learning_rate": 0.00014369999999999997,
"loss": 0.0556,
"step": 724
},
{
"epoch": 18.12,
"learning_rate": 0.0001434,
"loss": 0.059,
"step": 725
},
{
"epoch": 18.15,
"learning_rate": 0.00014309999999999998,
"loss": 0.0447,
"step": 726
},
{
"epoch": 18.18,
"learning_rate": 0.00014279999999999997,
"loss": 0.0572,
"step": 727
},
{
"epoch": 18.2,
"learning_rate": 0.0001425,
"loss": 0.0517,
"step": 728
},
{
"epoch": 18.23,
"learning_rate": 0.0001422,
"loss": 0.072,
"step": 729
},
{
"epoch": 18.25,
"learning_rate": 0.00014189999999999998,
"loss": 0.0518,
"step": 730
},
{
"epoch": 18.27,
"learning_rate": 0.00014159999999999997,
"loss": 0.0853,
"step": 731
},
{
"epoch": 18.3,
"learning_rate": 0.0001413,
"loss": 0.0657,
"step": 732
},
{
"epoch": 18.32,
"learning_rate": 0.00014099999999999998,
"loss": 0.0694,
"step": 733
},
{
"epoch": 18.35,
"learning_rate": 0.00014069999999999998,
"loss": 0.0775,
"step": 734
},
{
"epoch": 18.38,
"learning_rate": 0.0001404,
"loss": 0.0633,
"step": 735
},
{
"epoch": 18.4,
"learning_rate": 0.0001401,
"loss": 0.0446,
"step": 736
},
{
"epoch": 18.43,
"learning_rate": 0.00013979999999999998,
"loss": 0.051,
"step": 737
},
{
"epoch": 18.45,
"learning_rate": 0.0001395,
"loss": 0.0421,
"step": 738
},
{
"epoch": 18.48,
"learning_rate": 0.0001392,
"loss": 0.0387,
"step": 739
},
{
"epoch": 18.5,
"learning_rate": 0.0001389,
"loss": 0.0541,
"step": 740
},
{
"epoch": 18.52,
"learning_rate": 0.0001386,
"loss": 0.0609,
"step": 741
},
{
"epoch": 18.55,
"learning_rate": 0.0001383,
"loss": 0.0627,
"step": 742
},
{
"epoch": 18.57,
"learning_rate": 0.000138,
"loss": 0.0384,
"step": 743
},
{
"epoch": 18.6,
"learning_rate": 0.00013769999999999999,
"loss": 0.0392,
"step": 744
},
{
"epoch": 18.62,
"learning_rate": 0.0001374,
"loss": 0.0487,
"step": 745
},
{
"epoch": 18.65,
"learning_rate": 0.0001371,
"loss": 0.0811,
"step": 746
},
{
"epoch": 18.68,
"learning_rate": 0.0001368,
"loss": 0.0603,
"step": 747
},
{
"epoch": 18.7,
"learning_rate": 0.00013649999999999998,
"loss": 0.0502,
"step": 748
},
{
"epoch": 18.73,
"learning_rate": 0.0001362,
"loss": 0.0501,
"step": 749
},
{
"epoch": 18.75,
"learning_rate": 0.0001359,
"loss": 0.0344,
"step": 750
},
{
"epoch": 18.77,
"learning_rate": 0.0001356,
"loss": 0.053,
"step": 751
},
{
"epoch": 18.8,
"learning_rate": 0.00013529999999999998,
"loss": 0.0727,
"step": 752
},
{
"epoch": 18.82,
"learning_rate": 0.000135,
"loss": 0.0696,
"step": 753
},
{
"epoch": 18.85,
"learning_rate": 0.0001347,
"loss": 0.052,
"step": 754
},
{
"epoch": 18.88,
"learning_rate": 0.0001344,
"loss": 0.0613,
"step": 755
},
{
"epoch": 18.9,
"learning_rate": 0.00013409999999999998,
"loss": 0.0455,
"step": 756
},
{
"epoch": 18.93,
"learning_rate": 0.0001338,
"loss": 0.0411,
"step": 757
},
{
"epoch": 18.95,
"learning_rate": 0.0001335,
"loss": 0.0692,
"step": 758
},
{
"epoch": 18.98,
"learning_rate": 0.00013319999999999999,
"loss": 0.0478,
"step": 759
},
{
"epoch": 19.0,
"learning_rate": 0.00013289999999999998,
"loss": 0.0381,
"step": 760
},
{
"epoch": 19.02,
"learning_rate": 0.0001326,
"loss": 0.0501,
"step": 761
},
{
"epoch": 19.05,
"learning_rate": 0.0001323,
"loss": 0.0418,
"step": 762
},
{
"epoch": 19.07,
"learning_rate": 0.00013199999999999998,
"loss": 0.0448,
"step": 763
},
{
"epoch": 19.1,
"learning_rate": 0.00013169999999999998,
"loss": 0.0611,
"step": 764
},
{
"epoch": 19.12,
"learning_rate": 0.0001314,
"loss": 0.0431,
"step": 765
},
{
"epoch": 19.15,
"learning_rate": 0.0001311,
"loss": 0.0507,
"step": 766
},
{
"epoch": 19.18,
"learning_rate": 0.00013079999999999998,
"loss": 0.0578,
"step": 767
},
{
"epoch": 19.2,
"learning_rate": 0.0001305,
"loss": 0.0432,
"step": 768
},
{
"epoch": 19.23,
"learning_rate": 0.0001302,
"loss": 0.0516,
"step": 769
},
{
"epoch": 19.25,
"learning_rate": 0.00012989999999999999,
"loss": 0.0392,
"step": 770
},
{
"epoch": 19.27,
"learning_rate": 0.00012959999999999998,
"loss": 0.031,
"step": 771
},
{
"epoch": 19.3,
"learning_rate": 0.0001293,
"loss": 0.0646,
"step": 772
},
{
"epoch": 19.32,
"learning_rate": 0.000129,
"loss": 0.047,
"step": 773
},
{
"epoch": 19.35,
"learning_rate": 0.00012869999999999998,
"loss": 0.0586,
"step": 774
},
{
"epoch": 19.38,
"learning_rate": 0.00012839999999999998,
"loss": 0.0526,
"step": 775
},
{
"epoch": 19.4,
"learning_rate": 0.0001281,
"loss": 0.0422,
"step": 776
},
{
"epoch": 19.43,
"learning_rate": 0.0001278,
"loss": 0.0336,
"step": 777
},
{
"epoch": 19.45,
"learning_rate": 0.00012749999999999998,
"loss": 0.033,
"step": 778
},
{
"epoch": 19.48,
"learning_rate": 0.00012719999999999997,
"loss": 0.0593,
"step": 779
},
{
"epoch": 19.5,
"learning_rate": 0.0001269,
"loss": 0.0578,
"step": 780
},
{
"epoch": 19.52,
"learning_rate": 0.0001266,
"loss": 0.0588,
"step": 781
},
{
"epoch": 19.55,
"learning_rate": 0.00012629999999999998,
"loss": 0.0542,
"step": 782
},
{
"epoch": 19.57,
"learning_rate": 0.00012599999999999997,
"loss": 0.0485,
"step": 783
},
{
"epoch": 19.6,
"learning_rate": 0.0001257,
"loss": 0.0444,
"step": 784
},
{
"epoch": 19.62,
"learning_rate": 0.00012539999999999999,
"loss": 0.0407,
"step": 785
},
{
"epoch": 19.65,
"learning_rate": 0.00012509999999999998,
"loss": 0.0772,
"step": 786
},
{
"epoch": 19.68,
"learning_rate": 0.00012479999999999997,
"loss": 0.0577,
"step": 787
},
{
"epoch": 19.7,
"learning_rate": 0.0001245,
"loss": 0.0495,
"step": 788
},
{
"epoch": 19.73,
"learning_rate": 0.00012419999999999998,
"loss": 0.058,
"step": 789
},
{
"epoch": 19.75,
"learning_rate": 0.00012389999999999998,
"loss": 0.0481,
"step": 790
},
{
"epoch": 19.77,
"learning_rate": 0.0001236,
"loss": 0.0411,
"step": 791
},
{
"epoch": 19.8,
"learning_rate": 0.0001233,
"loss": 0.0591,
"step": 792
},
{
"epoch": 19.82,
"learning_rate": 0.00012299999999999998,
"loss": 0.0516,
"step": 793
},
{
"epoch": 19.85,
"learning_rate": 0.00012269999999999997,
"loss": 0.0542,
"step": 794
},
{
"epoch": 19.88,
"learning_rate": 0.0001224,
"loss": 0.0429,
"step": 795
},
{
"epoch": 19.9,
"learning_rate": 0.00012209999999999999,
"loss": 0.0949,
"step": 796
},
{
"epoch": 19.93,
"learning_rate": 0.00012179999999999999,
"loss": 0.0474,
"step": 797
},
{
"epoch": 19.95,
"learning_rate": 0.0001215,
"loss": 0.0506,
"step": 798
},
{
"epoch": 19.98,
"learning_rate": 0.00012119999999999999,
"loss": 0.0486,
"step": 799
},
{
"epoch": 20.0,
"learning_rate": 0.0001209,
"loss": 0.0334,
"step": 800
},
{
"epoch": 20.0,
"eval_cer": 0.1542545408941747,
"eval_loss": 0.5741926431655884,
"eval_runtime": 8.0485,
"eval_samples_per_second": 48.953,
"eval_steps_per_second": 6.212,
"eval_wer": 0.3666129232295672,
"step": 800
},
{
"epoch": 20.02,
"learning_rate": 0.00012059999999999999,
"loss": 0.0611,
"step": 801
},
{
"epoch": 20.05,
"learning_rate": 0.0001203,
"loss": 0.0542,
"step": 802
},
{
"epoch": 20.07,
"learning_rate": 0.00011999999999999999,
"loss": 0.0351,
"step": 803
},
{
"epoch": 20.1,
"learning_rate": 0.0001197,
"loss": 0.038,
"step": 804
},
{
"epoch": 20.12,
"learning_rate": 0.0001194,
"loss": 0.0368,
"step": 805
},
{
"epoch": 20.15,
"learning_rate": 0.0001191,
"loss": 0.0383,
"step": 806
},
{
"epoch": 20.18,
"learning_rate": 0.0001188,
"loss": 0.0456,
"step": 807
},
{
"epoch": 20.2,
"learning_rate": 0.0001185,
"loss": 0.0572,
"step": 808
},
{
"epoch": 20.23,
"learning_rate": 0.0001182,
"loss": 0.0488,
"step": 809
},
{
"epoch": 20.25,
"learning_rate": 0.00011789999999999999,
"loss": 0.0364,
"step": 810
},
{
"epoch": 20.27,
"learning_rate": 0.0001176,
"loss": 0.0516,
"step": 811
},
{
"epoch": 20.3,
"learning_rate": 0.00011729999999999999,
"loss": 0.0321,
"step": 812
},
{
"epoch": 20.32,
"learning_rate": 0.000117,
"loss": 0.0685,
"step": 813
},
{
"epoch": 20.35,
"learning_rate": 0.00011669999999999999,
"loss": 0.0497,
"step": 814
},
{
"epoch": 20.38,
"learning_rate": 0.0001164,
"loss": 0.0524,
"step": 815
},
{
"epoch": 20.4,
"learning_rate": 0.00011609999999999999,
"loss": 0.051,
"step": 816
},
{
"epoch": 20.43,
"learning_rate": 0.0001158,
"loss": 0.0449,
"step": 817
},
{
"epoch": 20.45,
"learning_rate": 0.00011549999999999999,
"loss": 0.0357,
"step": 818
},
{
"epoch": 20.48,
"learning_rate": 0.0001152,
"loss": 0.0388,
"step": 819
},
{
"epoch": 20.5,
"learning_rate": 0.00011489999999999999,
"loss": 0.0591,
"step": 820
},
{
"epoch": 20.52,
"learning_rate": 0.0001146,
"loss": 0.0579,
"step": 821
},
{
"epoch": 20.55,
"learning_rate": 0.00011429999999999999,
"loss": 0.0501,
"step": 822
},
{
"epoch": 20.57,
"learning_rate": 0.00011399999999999999,
"loss": 0.0425,
"step": 823
},
{
"epoch": 20.6,
"learning_rate": 0.00011369999999999999,
"loss": 0.0877,
"step": 824
},
{
"epoch": 20.62,
"learning_rate": 0.00011339999999999999,
"loss": 0.0333,
"step": 825
},
{
"epoch": 20.65,
"learning_rate": 0.00011309999999999998,
"loss": 0.0639,
"step": 826
},
{
"epoch": 20.68,
"learning_rate": 0.00011279999999999999,
"loss": 0.0441,
"step": 827
},
{
"epoch": 20.7,
"learning_rate": 0.0001125,
"loss": 0.0525,
"step": 828
},
{
"epoch": 20.73,
"learning_rate": 0.00011219999999999999,
"loss": 0.027,
"step": 829
},
{
"epoch": 20.75,
"learning_rate": 0.0001119,
"loss": 0.0337,
"step": 830
},
{
"epoch": 20.77,
"learning_rate": 0.00011159999999999999,
"loss": 0.0463,
"step": 831
},
{
"epoch": 20.8,
"learning_rate": 0.0001113,
"loss": 0.0633,
"step": 832
},
{
"epoch": 20.82,
"learning_rate": 0.00011099999999999999,
"loss": 0.0476,
"step": 833
},
{
"epoch": 20.85,
"learning_rate": 0.0001107,
"loss": 0.0453,
"step": 834
},
{
"epoch": 20.88,
"learning_rate": 0.00011039999999999999,
"loss": 0.0341,
"step": 835
},
{
"epoch": 20.9,
"learning_rate": 0.00011009999999999999,
"loss": 0.0324,
"step": 836
},
{
"epoch": 20.93,
"learning_rate": 0.00010979999999999999,
"loss": 0.0343,
"step": 837
},
{
"epoch": 20.95,
"learning_rate": 0.00010949999999999999,
"loss": 0.0429,
"step": 838
},
{
"epoch": 20.98,
"learning_rate": 0.00010919999999999998,
"loss": 0.0424,
"step": 839
},
{
"epoch": 21.0,
"learning_rate": 0.00010889999999999999,
"loss": 0.0349,
"step": 840
},
{
"epoch": 21.02,
"learning_rate": 0.00010859999999999998,
"loss": 0.0502,
"step": 841
},
{
"epoch": 21.05,
"learning_rate": 0.00010829999999999999,
"loss": 0.0455,
"step": 842
},
{
"epoch": 21.07,
"learning_rate": 0.00010799999999999998,
"loss": 0.0431,
"step": 843
},
{
"epoch": 21.1,
"learning_rate": 0.00010769999999999999,
"loss": 0.0334,
"step": 844
},
{
"epoch": 21.12,
"learning_rate": 0.00010739999999999998,
"loss": 0.0349,
"step": 845
},
{
"epoch": 21.15,
"learning_rate": 0.00010709999999999999,
"loss": 0.0338,
"step": 846
},
{
"epoch": 21.18,
"learning_rate": 0.00010679999999999998,
"loss": 0.0405,
"step": 847
},
{
"epoch": 21.2,
"learning_rate": 0.00010649999999999999,
"loss": 0.0435,
"step": 848
},
{
"epoch": 21.23,
"learning_rate": 0.00010619999999999998,
"loss": 0.0363,
"step": 849
},
{
"epoch": 21.25,
"learning_rate": 0.00010589999999999999,
"loss": 0.044,
"step": 850
},
{
"epoch": 21.27,
"learning_rate": 0.00010559999999999998,
"loss": 0.0276,
"step": 851
},
{
"epoch": 21.3,
"learning_rate": 0.00010529999999999998,
"loss": 0.0397,
"step": 852
},
{
"epoch": 21.32,
"learning_rate": 0.00010499999999999999,
"loss": 0.0479,
"step": 853
},
{
"epoch": 21.35,
"learning_rate": 0.00010469999999999998,
"loss": 0.052,
"step": 854
},
{
"epoch": 21.38,
"learning_rate": 0.00010439999999999999,
"loss": 0.0471,
"step": 855
},
{
"epoch": 21.4,
"learning_rate": 0.00010409999999999998,
"loss": 0.042,
"step": 856
},
{
"epoch": 21.43,
"learning_rate": 0.00010379999999999999,
"loss": 0.0446,
"step": 857
},
{
"epoch": 21.45,
"learning_rate": 0.00010349999999999998,
"loss": 0.0357,
"step": 858
},
{
"epoch": 21.48,
"learning_rate": 0.00010319999999999999,
"loss": 0.0435,
"step": 859
},
{
"epoch": 21.5,
"learning_rate": 0.0001029,
"loss": 0.0347,
"step": 860
},
{
"epoch": 21.52,
"learning_rate": 0.0001026,
"loss": 0.0483,
"step": 861
},
{
"epoch": 21.55,
"learning_rate": 0.00010229999999999999,
"loss": 0.04,
"step": 862
},
{
"epoch": 21.57,
"learning_rate": 0.000102,
"loss": 0.0388,
"step": 863
},
{
"epoch": 21.6,
"learning_rate": 0.00010169999999999999,
"loss": 0.0761,
"step": 864
},
{
"epoch": 21.62,
"learning_rate": 0.0001014,
"loss": 0.0354,
"step": 865
},
{
"epoch": 21.65,
"learning_rate": 0.0001011,
"loss": 0.0637,
"step": 866
},
{
"epoch": 21.68,
"learning_rate": 0.0001008,
"loss": 0.0539,
"step": 867
},
{
"epoch": 21.7,
"learning_rate": 0.0001005,
"loss": 0.0405,
"step": 868
},
{
"epoch": 21.73,
"learning_rate": 0.0001002,
"loss": 0.0382,
"step": 869
},
{
"epoch": 21.75,
"learning_rate": 9.99e-05,
"loss": 0.0454,
"step": 870
},
{
"epoch": 21.77,
"learning_rate": 9.96e-05,
"loss": 0.0416,
"step": 871
},
{
"epoch": 21.8,
"learning_rate": 9.93e-05,
"loss": 0.0465,
"step": 872
},
{
"epoch": 21.82,
"learning_rate": 9.9e-05,
"loss": 0.0528,
"step": 873
},
{
"epoch": 21.85,
"learning_rate": 9.87e-05,
"loss": 0.0373,
"step": 874
},
{
"epoch": 21.88,
"learning_rate": 9.839999999999999e-05,
"loss": 0.0313,
"step": 875
},
{
"epoch": 21.9,
"learning_rate": 9.81e-05,
"loss": 0.0317,
"step": 876
},
{
"epoch": 21.93,
"learning_rate": 9.779999999999999e-05,
"loss": 0.0425,
"step": 877
},
{
"epoch": 21.95,
"learning_rate": 9.75e-05,
"loss": 0.0425,
"step": 878
},
{
"epoch": 21.98,
"learning_rate": 9.719999999999999e-05,
"loss": 0.039,
"step": 879
},
{
"epoch": 22.0,
"learning_rate": 9.69e-05,
"loss": 0.0412,
"step": 880
},
{
"epoch": 22.02,
"learning_rate": 9.659999999999999e-05,
"loss": 0.0442,
"step": 881
},
{
"epoch": 22.05,
"learning_rate": 9.63e-05,
"loss": 0.0302,
"step": 882
},
{
"epoch": 22.07,
"learning_rate": 9.599999999999999e-05,
"loss": 0.0439,
"step": 883
},
{
"epoch": 22.1,
"learning_rate": 9.57e-05,
"loss": 0.0418,
"step": 884
},
{
"epoch": 22.12,
"learning_rate": 9.539999999999999e-05,
"loss": 0.0347,
"step": 885
},
{
"epoch": 22.15,
"learning_rate": 9.51e-05,
"loss": 0.0296,
"step": 886
},
{
"epoch": 22.18,
"learning_rate": 9.479999999999999e-05,
"loss": 0.0289,
"step": 887
},
{
"epoch": 22.2,
"learning_rate": 9.449999999999999e-05,
"loss": 0.0416,
"step": 888
},
{
"epoch": 22.23,
"learning_rate": 9.419999999999999e-05,
"loss": 0.042,
"step": 889
},
{
"epoch": 22.25,
"learning_rate": 9.389999999999999e-05,
"loss": 0.0307,
"step": 890
},
{
"epoch": 22.27,
"learning_rate": 9.36e-05,
"loss": 0.0284,
"step": 891
},
{
"epoch": 22.3,
"learning_rate": 9.329999999999999e-05,
"loss": 0.0357,
"step": 892
},
{
"epoch": 22.32,
"learning_rate": 9.3e-05,
"loss": 0.0571,
"step": 893
},
{
"epoch": 22.35,
"learning_rate": 9.269999999999999e-05,
"loss": 0.0481,
"step": 894
},
{
"epoch": 22.38,
"learning_rate": 9.24e-05,
"loss": 0.0464,
"step": 895
},
{
"epoch": 22.4,
"learning_rate": 9.209999999999999e-05,
"loss": 0.048,
"step": 896
},
{
"epoch": 22.43,
"learning_rate": 9.18e-05,
"loss": 0.0243,
"step": 897
},
{
"epoch": 22.45,
"learning_rate": 9.149999999999999e-05,
"loss": 0.0352,
"step": 898
},
{
"epoch": 22.48,
"learning_rate": 9.12e-05,
"loss": 0.0281,
"step": 899
},
{
"epoch": 22.5,
"learning_rate": 9.089999999999999e-05,
"loss": 0.0448,
"step": 900
},
{
"epoch": 22.52,
"learning_rate": 9.059999999999999e-05,
"loss": 0.0517,
"step": 901
},
{
"epoch": 22.55,
"learning_rate": 9.029999999999999e-05,
"loss": 0.0309,
"step": 902
},
{
"epoch": 22.57,
"learning_rate": 8.999999999999999e-05,
"loss": 0.0331,
"step": 903
},
{
"epoch": 22.6,
"learning_rate": 8.969999999999998e-05,
"loss": 0.0236,
"step": 904
},
{
"epoch": 22.62,
"learning_rate": 8.939999999999999e-05,
"loss": 0.0319,
"step": 905
},
{
"epoch": 22.65,
"learning_rate": 8.909999999999998e-05,
"loss": 0.0605,
"step": 906
},
{
"epoch": 22.68,
"learning_rate": 8.879999999999999e-05,
"loss": 0.0381,
"step": 907
},
{
"epoch": 22.7,
"learning_rate": 8.849999999999998e-05,
"loss": 0.029,
"step": 908
},
{
"epoch": 22.73,
"learning_rate": 8.819999999999999e-05,
"loss": 0.0186,
"step": 909
},
{
"epoch": 22.75,
"learning_rate": 8.789999999999998e-05,
"loss": 0.0836,
"step": 910
},
{
"epoch": 22.77,
"learning_rate": 8.759999999999999e-05,
"loss": 0.0171,
"step": 911
},
{
"epoch": 22.8,
"learning_rate": 8.729999999999998e-05,
"loss": 0.0451,
"step": 912
},
{
"epoch": 22.82,
"learning_rate": 8.699999999999999e-05,
"loss": 0.0371,
"step": 913
},
{
"epoch": 22.85,
"learning_rate": 8.669999999999998e-05,
"loss": 0.0378,
"step": 914
},
{
"epoch": 22.88,
"learning_rate": 8.639999999999999e-05,
"loss": 0.0365,
"step": 915
},
{
"epoch": 22.9,
"learning_rate": 8.609999999999999e-05,
"loss": 0.0401,
"step": 916
},
{
"epoch": 22.93,
"learning_rate": 8.579999999999998e-05,
"loss": 0.0341,
"step": 917
},
{
"epoch": 22.95,
"learning_rate": 8.549999999999999e-05,
"loss": 0.0496,
"step": 918
},
{
"epoch": 22.98,
"learning_rate": 8.519999999999998e-05,
"loss": 0.0266,
"step": 919
},
{
"epoch": 23.0,
"learning_rate": 8.489999999999999e-05,
"loss": 0.0281,
"step": 920
},
{
"epoch": 23.02,
"learning_rate": 8.459999999999998e-05,
"loss": 0.0471,
"step": 921
},
{
"epoch": 23.05,
"learning_rate": 8.43e-05,
"loss": 0.0524,
"step": 922
},
{
"epoch": 23.07,
"learning_rate": 8.4e-05,
"loss": 0.0387,
"step": 923
},
{
"epoch": 23.1,
"learning_rate": 8.37e-05,
"loss": 0.025,
"step": 924
},
{
"epoch": 23.12,
"learning_rate": 8.34e-05,
"loss": 0.0727,
"step": 925
},
{
"epoch": 23.15,
"learning_rate": 8.31e-05,
"loss": 0.0305,
"step": 926
},
{
"epoch": 23.18,
"learning_rate": 8.28e-05,
"loss": 0.0447,
"step": 927
},
{
"epoch": 23.2,
"learning_rate": 8.25e-05,
"loss": 0.0288,
"step": 928
},
{
"epoch": 23.23,
"learning_rate": 8.22e-05,
"loss": 0.0482,
"step": 929
},
{
"epoch": 23.25,
"learning_rate": 8.19e-05,
"loss": 0.0275,
"step": 930
},
{
"epoch": 23.27,
"learning_rate": 8.16e-05,
"loss": 0.0318,
"step": 931
},
{
"epoch": 23.3,
"learning_rate": 8.13e-05,
"loss": 0.0379,
"step": 932
},
{
"epoch": 23.32,
"learning_rate": 8.1e-05,
"loss": 0.0359,
"step": 933
},
{
"epoch": 23.35,
"learning_rate": 8.07e-05,
"loss": 0.0482,
"step": 934
},
{
"epoch": 23.38,
"learning_rate": 8.04e-05,
"loss": 0.0353,
"step": 935
},
{
"epoch": 23.4,
"learning_rate": 8.01e-05,
"loss": 0.0247,
"step": 936
},
{
"epoch": 23.43,
"learning_rate": 7.98e-05,
"loss": 0.0307,
"step": 937
},
{
"epoch": 23.45,
"learning_rate": 7.95e-05,
"loss": 0.0249,
"step": 938
},
{
"epoch": 23.48,
"learning_rate": 7.92e-05,
"loss": 0.04,
"step": 939
},
{
"epoch": 23.5,
"learning_rate": 7.89e-05,
"loss": 0.0457,
"step": 940
},
{
"epoch": 23.52,
"learning_rate": 7.86e-05,
"loss": 0.0376,
"step": 941
},
{
"epoch": 23.55,
"learning_rate": 7.829999999999999e-05,
"loss": 0.0432,
"step": 942
},
{
"epoch": 23.57,
"learning_rate": 7.8e-05,
"loss": 0.0257,
"step": 943
},
{
"epoch": 23.6,
"learning_rate": 7.769999999999999e-05,
"loss": 0.033,
"step": 944
},
{
"epoch": 23.62,
"learning_rate": 7.74e-05,
"loss": 0.0274,
"step": 945
},
{
"epoch": 23.65,
"learning_rate": 7.709999999999999e-05,
"loss": 0.0402,
"step": 946
},
{
"epoch": 23.68,
"learning_rate": 7.68e-05,
"loss": 0.0332,
"step": 947
},
{
"epoch": 23.7,
"learning_rate": 7.649999999999999e-05,
"loss": 0.0505,
"step": 948
},
{
"epoch": 23.73,
"learning_rate": 7.62e-05,
"loss": 0.0271,
"step": 949
},
{
"epoch": 23.75,
"learning_rate": 7.589999999999999e-05,
"loss": 0.0333,
"step": 950
},
{
"epoch": 23.77,
"learning_rate": 7.56e-05,
"loss": 0.0571,
"step": 951
},
{
"epoch": 23.8,
"learning_rate": 7.529999999999999e-05,
"loss": 0.0302,
"step": 952
},
{
"epoch": 23.82,
"learning_rate": 7.5e-05,
"loss": 0.0364,
"step": 953
},
{
"epoch": 23.85,
"learning_rate": 7.47e-05,
"loss": 0.032,
"step": 954
},
{
"epoch": 23.88,
"learning_rate": 7.439999999999999e-05,
"loss": 0.0374,
"step": 955
},
{
"epoch": 23.9,
"learning_rate": 7.41e-05,
"loss": 0.0474,
"step": 956
},
{
"epoch": 23.93,
"learning_rate": 7.379999999999999e-05,
"loss": 0.023,
"step": 957
},
{
"epoch": 23.95,
"learning_rate": 7.35e-05,
"loss": 0.0455,
"step": 958
},
{
"epoch": 23.98,
"learning_rate": 7.319999999999999e-05,
"loss": 0.0369,
"step": 959
},
{
"epoch": 24.0,
"learning_rate": 7.29e-05,
"loss": 0.0272,
"step": 960
},
{
"epoch": 24.02,
"learning_rate": 7.259999999999999e-05,
"loss": 0.0546,
"step": 961
},
{
"epoch": 24.05,
"learning_rate": 7.23e-05,
"loss": 0.0427,
"step": 962
},
{
"epoch": 24.07,
"learning_rate": 7.199999999999999e-05,
"loss": 0.0256,
"step": 963
},
{
"epoch": 24.1,
"learning_rate": 7.17e-05,
"loss": 0.0342,
"step": 964
},
{
"epoch": 24.12,
"learning_rate": 7.139999999999999e-05,
"loss": 0.073,
"step": 965
},
{
"epoch": 24.15,
"learning_rate": 7.11e-05,
"loss": 0.0234,
"step": 966
},
{
"epoch": 24.18,
"learning_rate": 7.079999999999999e-05,
"loss": 0.022,
"step": 967
},
{
"epoch": 24.2,
"learning_rate": 7.049999999999999e-05,
"loss": 0.0254,
"step": 968
},
{
"epoch": 24.23,
"learning_rate": 7.02e-05,
"loss": 0.0418,
"step": 969
},
{
"epoch": 24.25,
"learning_rate": 6.989999999999999e-05,
"loss": 0.0323,
"step": 970
},
{
"epoch": 24.27,
"learning_rate": 6.96e-05,
"loss": 0.0475,
"step": 971
},
{
"epoch": 24.3,
"learning_rate": 6.93e-05,
"loss": 0.0288,
"step": 972
},
{
"epoch": 24.32,
"learning_rate": 6.9e-05,
"loss": 0.0438,
"step": 973
},
{
"epoch": 24.35,
"learning_rate": 6.87e-05,
"loss": 0.0412,
"step": 974
},
{
"epoch": 24.38,
"learning_rate": 6.84e-05,
"loss": 0.024,
"step": 975
},
{
"epoch": 24.4,
"learning_rate": 6.81e-05,
"loss": 0.0327,
"step": 976
},
{
"epoch": 24.43,
"learning_rate": 6.78e-05,
"loss": 0.0334,
"step": 977
},
{
"epoch": 24.45,
"learning_rate": 6.75e-05,
"loss": 0.0245,
"step": 978
},
{
"epoch": 24.48,
"learning_rate": 6.72e-05,
"loss": 0.0361,
"step": 979
},
{
"epoch": 24.5,
"learning_rate": 6.69e-05,
"loss": 0.0471,
"step": 980
},
{
"epoch": 24.52,
"learning_rate": 6.659999999999999e-05,
"loss": 0.0484,
"step": 981
},
{
"epoch": 24.55,
"learning_rate": 6.63e-05,
"loss": 0.0324,
"step": 982
},
{
"epoch": 24.57,
"learning_rate": 6.599999999999999e-05,
"loss": 0.0244,
"step": 983
},
{
"epoch": 24.6,
"learning_rate": 6.57e-05,
"loss": 0.0205,
"step": 984
},
{
"epoch": 24.62,
"learning_rate": 6.539999999999999e-05,
"loss": 0.0252,
"step": 985
},
{
"epoch": 24.65,
"learning_rate": 6.51e-05,
"loss": 0.0381,
"step": 986
},
{
"epoch": 24.68,
"learning_rate": 6.479999999999999e-05,
"loss": 0.0411,
"step": 987
},
{
"epoch": 24.7,
"learning_rate": 6.45e-05,
"loss": 0.0284,
"step": 988
},
{
"epoch": 24.73,
"learning_rate": 6.419999999999999e-05,
"loss": 0.0319,
"step": 989
},
{
"epoch": 24.75,
"learning_rate": 6.39e-05,
"loss": 0.0287,
"step": 990
},
{
"epoch": 24.77,
"learning_rate": 6.359999999999999e-05,
"loss": 0.0266,
"step": 991
},
{
"epoch": 24.8,
"learning_rate": 6.33e-05,
"loss": 0.0569,
"step": 992
},
{
"epoch": 24.82,
"learning_rate": 6.299999999999999e-05,
"loss": 0.0373,
"step": 993
},
{
"epoch": 24.85,
"learning_rate": 6.269999999999999e-05,
"loss": 0.0425,
"step": 994
},
{
"epoch": 24.88,
"learning_rate": 6.239999999999999e-05,
"loss": 0.0283,
"step": 995
},
{
"epoch": 24.9,
"learning_rate": 6.209999999999999e-05,
"loss": 0.0271,
"step": 996
},
{
"epoch": 24.93,
"learning_rate": 6.18e-05,
"loss": 0.0285,
"step": 997
},
{
"epoch": 24.95,
"learning_rate": 6.149999999999999e-05,
"loss": 0.0252,
"step": 998
},
{
"epoch": 24.98,
"learning_rate": 6.12e-05,
"loss": 0.03,
"step": 999
},
{
"epoch": 25.0,
"learning_rate": 6.0899999999999996e-05,
"loss": 0.0244,
"step": 1000
},
{
"epoch": 25.0,
"eval_cer": 0.14988608573877057,
"eval_loss": 0.5907031893730164,
"eval_runtime": 8.0782,
"eval_samples_per_second": 48.773,
"eval_steps_per_second": 6.189,
"eval_wer": 0.3545826615403696,
"step": 1000
},
{
"epoch": 25.02,
"learning_rate": 6.0599999999999996e-05,
"loss": 0.0456,
"step": 1001
},
{
"epoch": 25.05,
"learning_rate": 6.0299999999999995e-05,
"loss": 0.0379,
"step": 1002
},
{
"epoch": 25.07,
"learning_rate": 5.9999999999999995e-05,
"loss": 0.027,
"step": 1003
},
{
"epoch": 25.1,
"learning_rate": 5.97e-05,
"loss": 0.0222,
"step": 1004
},
{
"epoch": 25.12,
"learning_rate": 5.94e-05,
"loss": 0.0291,
"step": 1005
},
{
"epoch": 25.15,
"learning_rate": 5.91e-05,
"loss": 0.0478,
"step": 1006
},
{
"epoch": 25.18,
"learning_rate": 5.88e-05,
"loss": 0.0294,
"step": 1007
},
{
"epoch": 25.2,
"learning_rate": 5.85e-05,
"loss": 0.0354,
"step": 1008
},
{
"epoch": 25.23,
"learning_rate": 5.82e-05,
"loss": 0.032,
"step": 1009
},
{
"epoch": 25.25,
"learning_rate": 5.79e-05,
"loss": 0.026,
"step": 1010
},
{
"epoch": 25.27,
"learning_rate": 5.76e-05,
"loss": 0.0228,
"step": 1011
},
{
"epoch": 25.3,
"learning_rate": 5.73e-05,
"loss": 0.02,
"step": 1012
},
{
"epoch": 25.32,
"learning_rate": 5.6999999999999996e-05,
"loss": 0.0354,
"step": 1013
},
{
"epoch": 25.35,
"learning_rate": 5.6699999999999996e-05,
"loss": 0.0352,
"step": 1014
},
{
"epoch": 25.38,
"learning_rate": 5.6399999999999995e-05,
"loss": 0.03,
"step": 1015
},
{
"epoch": 25.4,
"learning_rate": 5.6099999999999995e-05,
"loss": 0.0297,
"step": 1016
},
{
"epoch": 25.43,
"learning_rate": 5.5799999999999994e-05,
"loss": 0.0663,
"step": 1017
},
{
"epoch": 25.45,
"learning_rate": 5.5499999999999994e-05,
"loss": 0.0171,
"step": 1018
},
{
"epoch": 25.48,
"learning_rate": 5.519999999999999e-05,
"loss": 0.0242,
"step": 1019
},
{
"epoch": 25.5,
"learning_rate": 5.489999999999999e-05,
"loss": 0.0336,
"step": 1020
},
{
"epoch": 25.52,
"learning_rate": 5.459999999999999e-05,
"loss": 0.0447,
"step": 1021
},
{
"epoch": 25.55,
"learning_rate": 5.429999999999999e-05,
"loss": 0.0291,
"step": 1022
},
{
"epoch": 25.57,
"learning_rate": 5.399999999999999e-05,
"loss": 0.0274,
"step": 1023
},
{
"epoch": 25.6,
"learning_rate": 5.369999999999999e-05,
"loss": 0.0158,
"step": 1024
},
{
"epoch": 25.62,
"learning_rate": 5.339999999999999e-05,
"loss": 0.0287,
"step": 1025
},
{
"epoch": 25.65,
"learning_rate": 5.309999999999999e-05,
"loss": 0.0417,
"step": 1026
},
{
"epoch": 25.68,
"learning_rate": 5.279999999999999e-05,
"loss": 0.0328,
"step": 1027
},
{
"epoch": 25.7,
"learning_rate": 5.2499999999999995e-05,
"loss": 0.0432,
"step": 1028
},
{
"epoch": 25.73,
"learning_rate": 5.2199999999999995e-05,
"loss": 0.0387,
"step": 1029
},
{
"epoch": 25.75,
"learning_rate": 5.1899999999999994e-05,
"loss": 0.0274,
"step": 1030
},
{
"epoch": 25.77,
"learning_rate": 5.1599999999999994e-05,
"loss": 0.0265,
"step": 1031
},
{
"epoch": 25.8,
"learning_rate": 5.13e-05,
"loss": 0.0396,
"step": 1032
},
{
"epoch": 25.82,
"learning_rate": 5.1e-05,
"loss": 0.0324,
"step": 1033
},
{
"epoch": 25.85,
"learning_rate": 5.07e-05,
"loss": 0.0518,
"step": 1034
},
{
"epoch": 25.88,
"learning_rate": 5.04e-05,
"loss": 0.0379,
"step": 1035
},
{
"epoch": 25.9,
"learning_rate": 5.01e-05,
"loss": 0.0213,
"step": 1036
},
{
"epoch": 25.93,
"learning_rate": 4.98e-05,
"loss": 0.0329,
"step": 1037
},
{
"epoch": 25.95,
"learning_rate": 4.95e-05,
"loss": 0.0301,
"step": 1038
},
{
"epoch": 25.98,
"learning_rate": 4.9199999999999997e-05,
"loss": 0.0327,
"step": 1039
},
{
"epoch": 26.0,
"learning_rate": 4.8899999999999996e-05,
"loss": 0.0362,
"step": 1040
},
{
"epoch": 26.02,
"learning_rate": 4.8599999999999995e-05,
"loss": 0.0377,
"step": 1041
},
{
"epoch": 26.05,
"learning_rate": 4.8299999999999995e-05,
"loss": 0.0348,
"step": 1042
},
{
"epoch": 26.07,
"learning_rate": 4.7999999999999994e-05,
"loss": 0.024,
"step": 1043
},
{
"epoch": 26.1,
"learning_rate": 4.7699999999999994e-05,
"loss": 0.029,
"step": 1044
},
{
"epoch": 26.12,
"learning_rate": 4.7399999999999993e-05,
"loss": 0.0321,
"step": 1045
},
{
"epoch": 26.15,
"learning_rate": 4.709999999999999e-05,
"loss": 0.0317,
"step": 1046
},
{
"epoch": 26.18,
"learning_rate": 4.68e-05,
"loss": 0.034,
"step": 1047
},
{
"epoch": 26.2,
"learning_rate": 4.65e-05,
"loss": 0.0317,
"step": 1048
},
{
"epoch": 26.23,
"learning_rate": 4.62e-05,
"loss": 0.0316,
"step": 1049
},
{
"epoch": 26.25,
"learning_rate": 4.59e-05,
"loss": 0.0328,
"step": 1050
},
{
"epoch": 26.27,
"learning_rate": 4.56e-05,
"loss": 0.022,
"step": 1051
},
{
"epoch": 26.3,
"learning_rate": 4.5299999999999997e-05,
"loss": 0.029,
"step": 1052
},
{
"epoch": 26.32,
"learning_rate": 4.4999999999999996e-05,
"loss": 0.0247,
"step": 1053
},
{
"epoch": 26.35,
"learning_rate": 4.4699999999999996e-05,
"loss": 0.034,
"step": 1054
},
{
"epoch": 26.38,
"learning_rate": 4.4399999999999995e-05,
"loss": 0.0331,
"step": 1055
},
{
"epoch": 26.4,
"learning_rate": 4.4099999999999995e-05,
"loss": 0.0244,
"step": 1056
},
{
"epoch": 26.43,
"learning_rate": 4.3799999999999994e-05,
"loss": 0.0177,
"step": 1057
},
{
"epoch": 26.45,
"learning_rate": 4.3499999999999993e-05,
"loss": 0.0274,
"step": 1058
},
{
"epoch": 26.48,
"learning_rate": 4.319999999999999e-05,
"loss": 0.0283,
"step": 1059
},
{
"epoch": 26.5,
"learning_rate": 4.289999999999999e-05,
"loss": 0.0251,
"step": 1060
},
{
"epoch": 26.52,
"learning_rate": 4.259999999999999e-05,
"loss": 0.04,
"step": 1061
},
{
"epoch": 26.55,
"learning_rate": 4.229999999999999e-05,
"loss": 0.0316,
"step": 1062
},
{
"epoch": 26.57,
"learning_rate": 4.2e-05,
"loss": 0.0664,
"step": 1063
},
{
"epoch": 26.6,
"learning_rate": 4.17e-05,
"loss": 0.0174,
"step": 1064
},
{
"epoch": 26.62,
"learning_rate": 4.14e-05,
"loss": 0.0242,
"step": 1065
},
{
"epoch": 26.65,
"learning_rate": 4.11e-05,
"loss": 0.0354,
"step": 1066
},
{
"epoch": 26.68,
"learning_rate": 4.08e-05,
"loss": 0.0393,
"step": 1067
},
{
"epoch": 26.7,
"learning_rate": 4.05e-05,
"loss": 0.034,
"step": 1068
},
{
"epoch": 26.73,
"learning_rate": 4.02e-05,
"loss": 0.0311,
"step": 1069
},
{
"epoch": 26.75,
"learning_rate": 3.99e-05,
"loss": 0.0143,
"step": 1070
},
{
"epoch": 26.77,
"learning_rate": 3.96e-05,
"loss": 0.0208,
"step": 1071
},
{
"epoch": 26.8,
"learning_rate": 3.93e-05,
"loss": 0.0365,
"step": 1072
},
{
"epoch": 26.82,
"learning_rate": 3.9e-05,
"loss": 0.0241,
"step": 1073
},
{
"epoch": 26.85,
"learning_rate": 3.87e-05,
"loss": 0.0295,
"step": 1074
},
{
"epoch": 26.88,
"learning_rate": 3.84e-05,
"loss": 0.0221,
"step": 1075
},
{
"epoch": 26.9,
"learning_rate": 3.81e-05,
"loss": 0.0187,
"step": 1076
},
{
"epoch": 26.93,
"learning_rate": 3.78e-05,
"loss": 0.0197,
"step": 1077
},
{
"epoch": 26.95,
"learning_rate": 3.75e-05,
"loss": 0.0349,
"step": 1078
},
{
"epoch": 26.98,
"learning_rate": 3.7199999999999996e-05,
"loss": 0.0292,
"step": 1079
},
{
"epoch": 27.0,
"learning_rate": 3.6899999999999996e-05,
"loss": 0.0175,
"step": 1080
},
{
"epoch": 27.02,
"learning_rate": 3.6599999999999995e-05,
"loss": 0.029,
"step": 1081
},
{
"epoch": 27.05,
"learning_rate": 3.6299999999999995e-05,
"loss": 0.027,
"step": 1082
},
{
"epoch": 27.07,
"learning_rate": 3.5999999999999994e-05,
"loss": 0.0308,
"step": 1083
},
{
"epoch": 27.1,
"learning_rate": 3.5699999999999994e-05,
"loss": 0.0293,
"step": 1084
},
{
"epoch": 27.12,
"learning_rate": 3.539999999999999e-05,
"loss": 0.0198,
"step": 1085
},
{
"epoch": 27.15,
"learning_rate": 3.51e-05,
"loss": 0.0229,
"step": 1086
},
{
"epoch": 27.18,
"learning_rate": 3.48e-05,
"loss": 0.0391,
"step": 1087
},
{
"epoch": 27.2,
"learning_rate": 3.45e-05,
"loss": 0.0461,
"step": 1088
},
{
"epoch": 27.23,
"learning_rate": 3.42e-05,
"loss": 0.0265,
"step": 1089
},
{
"epoch": 27.25,
"learning_rate": 3.39e-05,
"loss": 0.0203,
"step": 1090
},
{
"epoch": 27.27,
"learning_rate": 3.36e-05,
"loss": 0.0225,
"step": 1091
},
{
"epoch": 27.3,
"learning_rate": 3.3299999999999996e-05,
"loss": 0.016,
"step": 1092
},
{
"epoch": 27.32,
"learning_rate": 3.2999999999999996e-05,
"loss": 0.0387,
"step": 1093
},
{
"epoch": 27.35,
"learning_rate": 3.2699999999999995e-05,
"loss": 0.0278,
"step": 1094
},
{
"epoch": 27.38,
"learning_rate": 3.2399999999999995e-05,
"loss": 0.0277,
"step": 1095
},
{
"epoch": 27.4,
"learning_rate": 3.2099999999999994e-05,
"loss": 0.0286,
"step": 1096
},
{
"epoch": 27.43,
"learning_rate": 3.1799999999999994e-05,
"loss": 0.023,
"step": 1097
},
{
"epoch": 27.45,
"learning_rate": 3.149999999999999e-05,
"loss": 0.0175,
"step": 1098
},
{
"epoch": 27.48,
"learning_rate": 3.119999999999999e-05,
"loss": 0.0193,
"step": 1099
},
{
"epoch": 27.5,
"learning_rate": 3.09e-05,
"loss": 0.0243,
"step": 1100
},
{
"epoch": 27.52,
"learning_rate": 3.06e-05,
"loss": 0.0251,
"step": 1101
},
{
"epoch": 27.55,
"learning_rate": 3.0299999999999998e-05,
"loss": 0.0207,
"step": 1102
},
{
"epoch": 27.57,
"learning_rate": 2.9999999999999997e-05,
"loss": 0.0227,
"step": 1103
},
{
"epoch": 27.6,
"learning_rate": 2.97e-05,
"loss": 0.0694,
"step": 1104
},
{
"epoch": 27.62,
"learning_rate": 2.94e-05,
"loss": 0.0346,
"step": 1105
},
{
"epoch": 27.65,
"learning_rate": 2.91e-05,
"loss": 0.0373,
"step": 1106
},
{
"epoch": 27.68,
"learning_rate": 2.88e-05,
"loss": 0.0259,
"step": 1107
},
{
"epoch": 27.7,
"learning_rate": 2.8499999999999998e-05,
"loss": 0.0307,
"step": 1108
},
{
"epoch": 27.73,
"learning_rate": 2.8199999999999998e-05,
"loss": 0.0229,
"step": 1109
},
{
"epoch": 27.75,
"learning_rate": 2.7899999999999997e-05,
"loss": 0.0202,
"step": 1110
},
{
"epoch": 27.77,
"learning_rate": 2.7599999999999997e-05,
"loss": 0.0231,
"step": 1111
},
{
"epoch": 27.8,
"learning_rate": 2.7299999999999996e-05,
"loss": 0.0244,
"step": 1112
},
{
"epoch": 27.82,
"learning_rate": 2.6999999999999996e-05,
"loss": 0.0311,
"step": 1113
},
{
"epoch": 27.85,
"learning_rate": 2.6699999999999995e-05,
"loss": 0.0447,
"step": 1114
},
{
"epoch": 27.88,
"learning_rate": 2.6399999999999995e-05,
"loss": 0.0293,
"step": 1115
},
{
"epoch": 27.9,
"learning_rate": 2.6099999999999997e-05,
"loss": 0.0255,
"step": 1116
},
{
"epoch": 27.93,
"learning_rate": 2.5799999999999997e-05,
"loss": 0.021,
"step": 1117
},
{
"epoch": 27.95,
"learning_rate": 2.55e-05,
"loss": 0.0276,
"step": 1118
},
{
"epoch": 27.98,
"learning_rate": 2.52e-05,
"loss": 0.0358,
"step": 1119
},
{
"epoch": 28.0,
"learning_rate": 2.49e-05,
"loss": 0.0136,
"step": 1120
},
{
"epoch": 28.02,
"learning_rate": 2.4599999999999998e-05,
"loss": 0.0417,
"step": 1121
},
{
"epoch": 28.05,
"learning_rate": 2.4299999999999998e-05,
"loss": 0.0293,
"step": 1122
},
{
"epoch": 28.07,
"learning_rate": 2.3999999999999997e-05,
"loss": 0.0401,
"step": 1123
},
{
"epoch": 28.1,
"learning_rate": 2.3699999999999997e-05,
"loss": 0.0195,
"step": 1124
},
{
"epoch": 28.12,
"learning_rate": 2.34e-05,
"loss": 0.0179,
"step": 1125
},
{
"epoch": 28.15,
"learning_rate": 2.31e-05,
"loss": 0.0201,
"step": 1126
},
{
"epoch": 28.18,
"learning_rate": 2.28e-05,
"loss": 0.0269,
"step": 1127
},
{
"epoch": 28.2,
"learning_rate": 2.2499999999999998e-05,
"loss": 0.0343,
"step": 1128
},
{
"epoch": 28.23,
"learning_rate": 2.2199999999999998e-05,
"loss": 0.0269,
"step": 1129
},
{
"epoch": 28.25,
"learning_rate": 2.1899999999999997e-05,
"loss": 0.0197,
"step": 1130
},
{
"epoch": 28.27,
"learning_rate": 2.1599999999999996e-05,
"loss": 0.029,
"step": 1131
},
{
"epoch": 28.3,
"learning_rate": 2.1299999999999996e-05,
"loss": 0.027,
"step": 1132
},
{
"epoch": 28.32,
"learning_rate": 2.1e-05,
"loss": 0.028,
"step": 1133
},
{
"epoch": 28.35,
"learning_rate": 2.07e-05,
"loss": 0.0298,
"step": 1134
},
{
"epoch": 28.38,
"learning_rate": 2.04e-05,
"loss": 0.0328,
"step": 1135
},
{
"epoch": 28.4,
"learning_rate": 2.01e-05,
"loss": 0.0235,
"step": 1136
},
{
"epoch": 28.43,
"learning_rate": 1.98e-05,
"loss": 0.0159,
"step": 1137
},
{
"epoch": 28.45,
"learning_rate": 1.95e-05,
"loss": 0.0151,
"step": 1138
},
{
"epoch": 28.48,
"learning_rate": 1.92e-05,
"loss": 0.0227,
"step": 1139
},
{
"epoch": 28.5,
"learning_rate": 1.89e-05,
"loss": 0.0283,
"step": 1140
},
{
"epoch": 28.52,
"learning_rate": 1.8599999999999998e-05,
"loss": 0.0229,
"step": 1141
},
{
"epoch": 28.55,
"learning_rate": 1.8299999999999998e-05,
"loss": 0.0259,
"step": 1142
},
{
"epoch": 28.57,
"learning_rate": 1.7999999999999997e-05,
"loss": 0.0298,
"step": 1143
},
{
"epoch": 28.6,
"learning_rate": 1.7699999999999997e-05,
"loss": 0.0709,
"step": 1144
},
{
"epoch": 28.62,
"learning_rate": 1.74e-05,
"loss": 0.0279,
"step": 1145
},
{
"epoch": 28.65,
"learning_rate": 1.71e-05,
"loss": 0.0321,
"step": 1146
},
{
"epoch": 28.68,
"learning_rate": 1.68e-05,
"loss": 0.0268,
"step": 1147
},
{
"epoch": 28.7,
"learning_rate": 1.6499999999999998e-05,
"loss": 0.0341,
"step": 1148
},
{
"epoch": 28.73,
"learning_rate": 1.6199999999999997e-05,
"loss": 0.0224,
"step": 1149
},
{
"epoch": 28.75,
"learning_rate": 1.5899999999999997e-05,
"loss": 0.0206,
"step": 1150
},
{
"epoch": 28.77,
"learning_rate": 1.5599999999999996e-05,
"loss": 0.0299,
"step": 1151
},
{
"epoch": 28.8,
"learning_rate": 1.53e-05,
"loss": 0.0375,
"step": 1152
},
{
"epoch": 28.82,
"learning_rate": 1.4999999999999999e-05,
"loss": 0.0376,
"step": 1153
},
{
"epoch": 28.85,
"learning_rate": 1.47e-05,
"loss": 0.0286,
"step": 1154
},
{
"epoch": 28.88,
"learning_rate": 1.44e-05,
"loss": 0.0235,
"step": 1155
},
{
"epoch": 28.9,
"learning_rate": 1.4099999999999999e-05,
"loss": 0.021,
"step": 1156
},
{
"epoch": 28.93,
"learning_rate": 1.3799999999999998e-05,
"loss": 0.0225,
"step": 1157
},
{
"epoch": 28.95,
"learning_rate": 1.3499999999999998e-05,
"loss": 0.03,
"step": 1158
},
{
"epoch": 28.98,
"learning_rate": 1.3199999999999997e-05,
"loss": 0.0395,
"step": 1159
},
{
"epoch": 29.0,
"learning_rate": 1.2899999999999998e-05,
"loss": 0.0231,
"step": 1160
},
{
"epoch": 29.02,
"learning_rate": 1.26e-05,
"loss": 0.04,
"step": 1161
},
{
"epoch": 29.05,
"learning_rate": 1.2299999999999999e-05,
"loss": 0.0264,
"step": 1162
},
{
"epoch": 29.07,
"learning_rate": 1.1999999999999999e-05,
"loss": 0.0284,
"step": 1163
},
{
"epoch": 29.1,
"learning_rate": 1.17e-05,
"loss": 0.0262,
"step": 1164
},
{
"epoch": 29.12,
"learning_rate": 1.14e-05,
"loss": 0.0165,
"step": 1165
},
{
"epoch": 29.15,
"learning_rate": 1.1099999999999999e-05,
"loss": 0.016,
"step": 1166
},
{
"epoch": 29.18,
"learning_rate": 1.0799999999999998e-05,
"loss": 0.0307,
"step": 1167
},
{
"epoch": 29.2,
"learning_rate": 1.05e-05,
"loss": 0.0387,
"step": 1168
},
{
"epoch": 29.23,
"learning_rate": 1.02e-05,
"loss": 0.0272,
"step": 1169
},
{
"epoch": 29.25,
"learning_rate": 9.9e-06,
"loss": 0.0201,
"step": 1170
},
{
"epoch": 29.27,
"learning_rate": 9.6e-06,
"loss": 0.0246,
"step": 1171
},
{
"epoch": 29.3,
"learning_rate": 9.299999999999999e-06,
"loss": 0.0309,
"step": 1172
},
{
"epoch": 29.32,
"learning_rate": 8.999999999999999e-06,
"loss": 0.033,
"step": 1173
},
{
"epoch": 29.35,
"learning_rate": 8.7e-06,
"loss": 0.0267,
"step": 1174
},
{
"epoch": 29.38,
"learning_rate": 8.4e-06,
"loss": 0.0267,
"step": 1175
},
{
"epoch": 29.4,
"learning_rate": 8.099999999999999e-06,
"loss": 0.0255,
"step": 1176
},
{
"epoch": 29.43,
"learning_rate": 7.799999999999998e-06,
"loss": 0.0326,
"step": 1177
},
{
"epoch": 29.45,
"learning_rate": 7.499999999999999e-06,
"loss": 0.0309,
"step": 1178
},
{
"epoch": 29.48,
"learning_rate": 7.2e-06,
"loss": 0.0234,
"step": 1179
},
{
"epoch": 29.5,
"learning_rate": 6.899999999999999e-06,
"loss": 0.0288,
"step": 1180
},
{
"epoch": 29.52,
"learning_rate": 6.599999999999999e-06,
"loss": 0.0244,
"step": 1181
},
{
"epoch": 29.55,
"learning_rate": 6.3e-06,
"loss": 0.0282,
"step": 1182
},
{
"epoch": 29.57,
"learning_rate": 5.999999999999999e-06,
"loss": 0.0202,
"step": 1183
},
{
"epoch": 29.6,
"learning_rate": 5.7e-06,
"loss": 0.0657,
"step": 1184
},
{
"epoch": 29.62,
"learning_rate": 5.399999999999999e-06,
"loss": 0.0205,
"step": 1185
},
{
"epoch": 29.65,
"learning_rate": 5.1e-06,
"loss": 0.0422,
"step": 1186
},
{
"epoch": 29.68,
"learning_rate": 4.8e-06,
"loss": 0.0203,
"step": 1187
},
{
"epoch": 29.7,
"learning_rate": 4.499999999999999e-06,
"loss": 0.0197,
"step": 1188
},
{
"epoch": 29.73,
"learning_rate": 4.2e-06,
"loss": 0.032,
"step": 1189
},
{
"epoch": 29.75,
"learning_rate": 3.899999999999999e-06,
"loss": 0.0243,
"step": 1190
},
{
"epoch": 29.77,
"learning_rate": 3.6e-06,
"loss": 0.0162,
"step": 1191
},
{
"epoch": 29.8,
"learning_rate": 3.2999999999999993e-06,
"loss": 0.0353,
"step": 1192
},
{
"epoch": 29.82,
"learning_rate": 2.9999999999999997e-06,
"loss": 0.0329,
"step": 1193
},
{
"epoch": 29.85,
"learning_rate": 2.6999999999999996e-06,
"loss": 0.0159,
"step": 1194
},
{
"epoch": 29.88,
"learning_rate": 2.4e-06,
"loss": 0.0263,
"step": 1195
},
{
"epoch": 29.9,
"learning_rate": 2.1e-06,
"loss": 0.0211,
"step": 1196
},
{
"epoch": 29.93,
"learning_rate": 1.8e-06,
"loss": 0.0178,
"step": 1197
},
{
"epoch": 29.95,
"learning_rate": 1.4999999999999998e-06,
"loss": 0.0278,
"step": 1198
},
{
"epoch": 29.98,
"learning_rate": 1.2e-06,
"loss": 0.0283,
"step": 1199
},
{
"epoch": 30.0,
"learning_rate": 9e-07,
"loss": 0.0143,
"step": 1200
},
{
"epoch": 30.0,
"eval_cer": 0.14687624103839642,
"eval_loss": 0.5961307287216187,
"eval_runtime": 8.1271,
"eval_samples_per_second": 48.48,
"eval_steps_per_second": 6.152,
"eval_wer": 0.34602505270990946,
"step": 1200
},
{
"epoch": 30.0,
"step": 1200,
"total_flos": 2.747721152315628e+19,
"train_loss": 1.155013433122852,
"train_runtime": 1766.4943,
"train_samples_per_second": 43.306,
"train_steps_per_second": 0.679
}
],
"max_steps": 1200,
"num_train_epochs": 30,
"total_flos": 2.747721152315628e+19,
"trial_name": null,
"trial_params": null
}