End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 38.0,
|
| 3 |
+
"eval_loss": 0.3874845504760742,
|
| 4 |
+
"eval_runtime": 122.8953,
|
| 5 |
+
"eval_samples_per_second": 2.23,
|
| 6 |
+
"eval_steps_per_second": 0.146,
|
| 7 |
+
"eval_wer": 6.531481563650885,
|
| 8 |
+
"train_loss": 0.18918936033500358,
|
| 9 |
+
"train_runtime": 13916.8243,
|
| 10 |
+
"train_samples_per_second": 1.84,
|
| 11 |
+
"train_steps_per_second": 0.057
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 38.0,
|
| 3 |
+
"eval_loss": 0.3874845504760742,
|
| 4 |
+
"eval_runtime": 122.8953,
|
| 5 |
+
"eval_samples_per_second": 2.23,
|
| 6 |
+
"eval_steps_per_second": 0.146,
|
| 7 |
+
"eval_wer": 6.531481563650885
|
| 8 |
+
}
|
runs/Feb21_17-40-17_dc6bdc3f9200/events.out.tfevents.1677016051.dc6bdc3f9200.1.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:596a293a1c9fc45978238a419a3b5dcf698f1b4e1f36a1f3384acbe72bc80d14
|
| 3 |
+
size 406
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 38.0,
|
| 3 |
+
"train_loss": 0.18918936033500358,
|
| 4 |
+
"train_runtime": 13916.8243,
|
| 5 |
+
"train_samples_per_second": 1.84,
|
| 6 |
+
"train_steps_per_second": 0.057
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,649 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 6.531481563650885,
|
| 3 |
+
"best_model_checkpoint": "whisper-stakwork-crypto/checkpoint-150",
|
| 4 |
+
"epoch": 38.0025,
|
| 5 |
+
"global_step": 800,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 3.0000000000000004e-08,
|
| 13 |
+
"loss": 0.8208,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.03,
|
| 18 |
+
"learning_rate": 8e-08,
|
| 19 |
+
"loss": 0.8146,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.01,
|
| 24 |
+
"learning_rate": 1.3e-07,
|
| 25 |
+
"loss": 0.7898,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.02,
|
| 30 |
+
"learning_rate": 1.8e-07,
|
| 31 |
+
"loss": 0.7761,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 2.01,
|
| 36 |
+
"learning_rate": 2.3e-07,
|
| 37 |
+
"loss": 0.7348,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 2.01,
|
| 42 |
+
"eval_loss": 0.7377644777297974,
|
| 43 |
+
"eval_runtime": 118.0826,
|
| 44 |
+
"eval_samples_per_second": 2.32,
|
| 45 |
+
"eval_steps_per_second": 0.152,
|
| 46 |
+
"eval_wer": 9.30913608732307,
|
| 47 |
+
"step": 50
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 2.02,
|
| 51 |
+
"learning_rate": 2.8e-07,
|
| 52 |
+
"loss": 0.7129,
|
| 53 |
+
"step": 60
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 3.01,
|
| 57 |
+
"learning_rate": 3.3e-07,
|
| 58 |
+
"loss": 0.6506,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 3.02,
|
| 63 |
+
"learning_rate": 3.8000000000000007e-07,
|
| 64 |
+
"loss": 0.6193,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 4.01,
|
| 69 |
+
"learning_rate": 4.3e-07,
|
| 70 |
+
"loss": 0.574,
|
| 71 |
+
"step": 90
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 4.02,
|
| 75 |
+
"learning_rate": 4.800000000000001e-07,
|
| 76 |
+
"loss": 0.5276,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 4.02,
|
| 81 |
+
"eval_loss": 0.5290005207061768,
|
| 82 |
+
"eval_runtime": 121.5574,
|
| 83 |
+
"eval_samples_per_second": 2.254,
|
| 84 |
+
"eval_steps_per_second": 0.148,
|
| 85 |
+
"eval_wer": 8.612503882504326,
|
| 86 |
+
"step": 100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 5.01,
|
| 90 |
+
"learning_rate": 5.3e-07,
|
| 91 |
+
"loss": 0.4652,
|
| 92 |
+
"step": 110
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 5.02,
|
| 96 |
+
"learning_rate": 5.8e-07,
|
| 97 |
+
"loss": 0.4235,
|
| 98 |
+
"step": 120
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 6.0,
|
| 102 |
+
"learning_rate": 6.3e-07,
|
| 103 |
+
"loss": 0.3972,
|
| 104 |
+
"step": 130
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.02,
|
| 108 |
+
"learning_rate": 6.8e-07,
|
| 109 |
+
"loss": 0.3675,
|
| 110 |
+
"step": 140
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 7.0,
|
| 114 |
+
"learning_rate": 7.300000000000001e-07,
|
| 115 |
+
"loss": 0.3585,
|
| 116 |
+
"step": 150
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 7.0,
|
| 120 |
+
"eval_loss": 0.3874845504760742,
|
| 121 |
+
"eval_runtime": 130.7132,
|
| 122 |
+
"eval_samples_per_second": 2.096,
|
| 123 |
+
"eval_steps_per_second": 0.138,
|
| 124 |
+
"eval_wer": 6.531481563650885,
|
| 125 |
+
"step": 150
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 7.02,
|
| 129 |
+
"learning_rate": 7.8e-07,
|
| 130 |
+
"loss": 0.3306,
|
| 131 |
+
"step": 160
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 8.0,
|
| 135 |
+
"learning_rate": 8.3e-07,
|
| 136 |
+
"loss": 0.3255,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 8.02,
|
| 141 |
+
"learning_rate": 8.8e-07,
|
| 142 |
+
"loss": 0.307,
|
| 143 |
+
"step": 180
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 9.0,
|
| 147 |
+
"learning_rate": 9.3e-07,
|
| 148 |
+
"loss": 0.3035,
|
| 149 |
+
"step": 190
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 9.01,
|
| 153 |
+
"learning_rate": 9.8e-07,
|
| 154 |
+
"loss": 0.2924,
|
| 155 |
+
"step": 200
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 9.01,
|
| 159 |
+
"eval_loss": 0.35476040840148926,
|
| 160 |
+
"eval_runtime": 120.7053,
|
| 161 |
+
"eval_samples_per_second": 2.27,
|
| 162 |
+
"eval_steps_per_second": 0.149,
|
| 163 |
+
"eval_wer": 6.677907441096862,
|
| 164 |
+
"step": 200
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 9.03,
|
| 168 |
+
"learning_rate": 1.03e-06,
|
| 169 |
+
"loss": 0.2826,
|
| 170 |
+
"step": 210
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 10.01,
|
| 174 |
+
"learning_rate": 1.08e-06,
|
| 175 |
+
"loss": 0.267,
|
| 176 |
+
"step": 220
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 10.03,
|
| 180 |
+
"learning_rate": 1.13e-06,
|
| 181 |
+
"loss": 0.27,
|
| 182 |
+
"step": 230
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 11.01,
|
| 186 |
+
"learning_rate": 1.18e-06,
|
| 187 |
+
"loss": 0.2492,
|
| 188 |
+
"step": 240
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 11.02,
|
| 192 |
+
"learning_rate": 1.2299999999999999e-06,
|
| 193 |
+
"loss": 0.2506,
|
| 194 |
+
"step": 250
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 11.02,
|
| 198 |
+
"eval_loss": 0.33637768030166626,
|
| 199 |
+
"eval_runtime": 118.1569,
|
| 200 |
+
"eval_samples_per_second": 2.319,
|
| 201 |
+
"eval_steps_per_second": 0.152,
|
| 202 |
+
"eval_wer": 6.788836136131694,
|
| 203 |
+
"step": 250
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 12.01,
|
| 207 |
+
"learning_rate": 1.28e-06,
|
| 208 |
+
"loss": 0.2263,
|
| 209 |
+
"step": 260
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 12.02,
|
| 213 |
+
"learning_rate": 1.3300000000000002e-06,
|
| 214 |
+
"loss": 0.23,
|
| 215 |
+
"step": 270
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 13.01,
|
| 219 |
+
"learning_rate": 1.3800000000000001e-06,
|
| 220 |
+
"loss": 0.2102,
|
| 221 |
+
"step": 280
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 13.02,
|
| 225 |
+
"learning_rate": 1.43e-06,
|
| 226 |
+
"loss": 0.2093,
|
| 227 |
+
"step": 290
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 14.01,
|
| 231 |
+
"learning_rate": 1.48e-06,
|
| 232 |
+
"loss": 0.1946,
|
| 233 |
+
"step": 300
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 14.01,
|
| 237 |
+
"eval_loss": 0.32616764307022095,
|
| 238 |
+
"eval_runtime": 119.3365,
|
| 239 |
+
"eval_samples_per_second": 2.296,
|
| 240 |
+
"eval_steps_per_second": 0.151,
|
| 241 |
+
"eval_wer": 7.148245108044549,
|
| 242 |
+
"step": 300
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 14.02,
|
| 246 |
+
"learning_rate": 1.53e-06,
|
| 247 |
+
"loss": 0.1848,
|
| 248 |
+
"step": 310
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 15.01,
|
| 252 |
+
"learning_rate": 1.58e-06,
|
| 253 |
+
"loss": 0.1752,
|
| 254 |
+
"step": 320
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 15.02,
|
| 258 |
+
"learning_rate": 1.63e-06,
|
| 259 |
+
"loss": 0.1659,
|
| 260 |
+
"step": 330
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 16.0,
|
| 264 |
+
"learning_rate": 1.6800000000000002e-06,
|
| 265 |
+
"loss": 0.1579,
|
| 266 |
+
"step": 340
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 16.02,
|
| 270 |
+
"learning_rate": 1.73e-06,
|
| 271 |
+
"loss": 0.1411,
|
| 272 |
+
"step": 350
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 16.02,
|
| 276 |
+
"eval_loss": 0.33289629220962524,
|
| 277 |
+
"eval_runtime": 117.7955,
|
| 278 |
+
"eval_samples_per_second": 2.326,
|
| 279 |
+
"eval_steps_per_second": 0.153,
|
| 280 |
+
"eval_wer": 7.210365177264055,
|
| 281 |
+
"step": 350
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 17.0,
|
| 285 |
+
"learning_rate": 1.7800000000000001e-06,
|
| 286 |
+
"loss": 0.1374,
|
| 287 |
+
"step": 360
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 17.02,
|
| 291 |
+
"learning_rate": 1.83e-06,
|
| 292 |
+
"loss": 0.1223,
|
| 293 |
+
"step": 370
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 18.0,
|
| 297 |
+
"learning_rate": 1.8800000000000002e-06,
|
| 298 |
+
"loss": 0.1173,
|
| 299 |
+
"step": 380
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 18.02,
|
| 303 |
+
"learning_rate": 1.93e-06,
|
| 304 |
+
"loss": 0.1034,
|
| 305 |
+
"step": 390
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 19.0,
|
| 309 |
+
"learning_rate": 1.98e-06,
|
| 310 |
+
"loss": 0.1005,
|
| 311 |
+
"step": 400
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 19.0,
|
| 315 |
+
"eval_loss": 0.34216001629829407,
|
| 316 |
+
"eval_runtime": 121.3922,
|
| 317 |
+
"eval_samples_per_second": 2.257,
|
| 318 |
+
"eval_steps_per_second": 0.148,
|
| 319 |
+
"eval_wer": 7.556462705772729,
|
| 320 |
+
"step": 400
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 19.01,
|
| 324 |
+
"learning_rate": 2.03e-06,
|
| 325 |
+
"loss": 0.0894,
|
| 326 |
+
"step": 410
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 19.03,
|
| 330 |
+
"learning_rate": 2.08e-06,
|
| 331 |
+
"loss": 0.0786,
|
| 332 |
+
"step": 420
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 20.01,
|
| 336 |
+
"learning_rate": 2.13e-06,
|
| 337 |
+
"loss": 0.0694,
|
| 338 |
+
"step": 430
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 20.02,
|
| 342 |
+
"learning_rate": 2.1800000000000003e-06,
|
| 343 |
+
"loss": 0.066,
|
| 344 |
+
"step": 440
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 21.01,
|
| 348 |
+
"learning_rate": 2.23e-06,
|
| 349 |
+
"loss": 0.0535,
|
| 350 |
+
"step": 450
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 21.01,
|
| 354 |
+
"eval_loss": 0.35322892665863037,
|
| 355 |
+
"eval_runtime": 120.2632,
|
| 356 |
+
"eval_samples_per_second": 2.278,
|
| 357 |
+
"eval_steps_per_second": 0.15,
|
| 358 |
+
"eval_wer": 7.179305142654302,
|
| 359 |
+
"step": 450
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 21.02,
|
| 363 |
+
"learning_rate": 2.28e-06,
|
| 364 |
+
"loss": 0.0494,
|
| 365 |
+
"step": 460
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 22.01,
|
| 369 |
+
"learning_rate": 2.3299999999999997e-06,
|
| 370 |
+
"loss": 0.043,
|
| 371 |
+
"step": 470
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 22.02,
|
| 375 |
+
"learning_rate": 2.38e-06,
|
| 376 |
+
"loss": 0.0359,
|
| 377 |
+
"step": 480
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 23.01,
|
| 381 |
+
"learning_rate": 2.43e-06,
|
| 382 |
+
"loss": 0.0311,
|
| 383 |
+
"step": 490
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 23.02,
|
| 387 |
+
"learning_rate": 2.48e-06,
|
| 388 |
+
"loss": 0.0259,
|
| 389 |
+
"step": 500
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 23.02,
|
| 393 |
+
"eval_loss": 0.34564533829689026,
|
| 394 |
+
"eval_runtime": 119.3022,
|
| 395 |
+
"eval_samples_per_second": 2.297,
|
| 396 |
+
"eval_steps_per_second": 0.151,
|
| 397 |
+
"eval_wer": 7.512091227758797,
|
| 398 |
+
"step": 500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 24.01,
|
| 402 |
+
"learning_rate": 2.5300000000000003e-06,
|
| 403 |
+
"loss": 0.0222,
|
| 404 |
+
"step": 510
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 24.02,
|
| 408 |
+
"learning_rate": 2.58e-06,
|
| 409 |
+
"loss": 0.0186,
|
| 410 |
+
"step": 520
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 25.01,
|
| 414 |
+
"learning_rate": 2.6300000000000002e-06,
|
| 415 |
+
"loss": 0.0154,
|
| 416 |
+
"step": 530
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 25.02,
|
| 420 |
+
"learning_rate": 2.68e-06,
|
| 421 |
+
"loss": 0.0154,
|
| 422 |
+
"step": 540
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 26.0,
|
| 426 |
+
"learning_rate": 2.73e-06,
|
| 427 |
+
"loss": 0.0137,
|
| 428 |
+
"step": 550
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 26.0,
|
| 432 |
+
"eval_loss": 0.3586655259132385,
|
| 433 |
+
"eval_runtime": 121.9747,
|
| 434 |
+
"eval_samples_per_second": 2.246,
|
| 435 |
+
"eval_steps_per_second": 0.148,
|
| 436 |
+
"eval_wer": 7.654079957403381,
|
| 437 |
+
"step": 550
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 26.02,
|
| 441 |
+
"learning_rate": 2.78e-06,
|
| 442 |
+
"loss": 0.0121,
|
| 443 |
+
"step": 560
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 27.0,
|
| 447 |
+
"learning_rate": 2.83e-06,
|
| 448 |
+
"loss": 0.0113,
|
| 449 |
+
"step": 570
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 27.02,
|
| 453 |
+
"learning_rate": 2.88e-06,
|
| 454 |
+
"loss": 0.0103,
|
| 455 |
+
"step": 580
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 28.0,
|
| 459 |
+
"learning_rate": 2.9300000000000003e-06,
|
| 460 |
+
"loss": 0.0112,
|
| 461 |
+
"step": 590
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 28.02,
|
| 465 |
+
"learning_rate": 2.98e-06,
|
| 466 |
+
"loss": 0.0078,
|
| 467 |
+
"step": 600
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 28.02,
|
| 471 |
+
"eval_loss": 0.3590793013572693,
|
| 472 |
+
"eval_runtime": 125.6972,
|
| 473 |
+
"eval_samples_per_second": 2.18,
|
| 474 |
+
"eval_steps_per_second": 0.143,
|
| 475 |
+
"eval_wer": 7.352353906908639,
|
| 476 |
+
"step": 600
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 29.0,
|
| 480 |
+
"learning_rate": 2.91e-06,
|
| 481 |
+
"loss": 0.0074,
|
| 482 |
+
"step": 610
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 29.01,
|
| 486 |
+
"learning_rate": 2.7600000000000003e-06,
|
| 487 |
+
"loss": 0.0065,
|
| 488 |
+
"step": 620
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 29.03,
|
| 492 |
+
"learning_rate": 2.61e-06,
|
| 493 |
+
"loss": 0.0061,
|
| 494 |
+
"step": 630
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 30.01,
|
| 498 |
+
"learning_rate": 2.4599999999999997e-06,
|
| 499 |
+
"loss": 0.0053,
|
| 500 |
+
"step": 640
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 30.02,
|
| 504 |
+
"learning_rate": 2.31e-06,
|
| 505 |
+
"loss": 0.0041,
|
| 506 |
+
"step": 650
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 30.02,
|
| 510 |
+
"eval_loss": 0.3672121465206146,
|
| 511 |
+
"eval_runtime": 119.005,
|
| 512 |
+
"eval_samples_per_second": 2.302,
|
| 513 |
+
"eval_steps_per_second": 0.151,
|
| 514 |
+
"eval_wer": 7.303545281093313,
|
| 515 |
+
"step": 650
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 31.01,
|
| 519 |
+
"learning_rate": 2.16e-06,
|
| 520 |
+
"loss": 0.0035,
|
| 521 |
+
"step": 660
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 31.02,
|
| 525 |
+
"learning_rate": 2.0100000000000002e-06,
|
| 526 |
+
"loss": 0.0032,
|
| 527 |
+
"step": 670
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 32.01,
|
| 531 |
+
"learning_rate": 1.86e-06,
|
| 532 |
+
"loss": 0.0029,
|
| 533 |
+
"step": 680
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 32.02,
|
| 537 |
+
"learning_rate": 1.71e-06,
|
| 538 |
+
"loss": 0.0028,
|
| 539 |
+
"step": 690
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 33.01,
|
| 543 |
+
"learning_rate": 1.56e-06,
|
| 544 |
+
"loss": 0.0026,
|
| 545 |
+
"step": 700
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 33.01,
|
| 549 |
+
"eval_loss": 0.3961912989616394,
|
| 550 |
+
"eval_runtime": 131.1922,
|
| 551 |
+
"eval_samples_per_second": 2.089,
|
| 552 |
+
"eval_steps_per_second": 0.137,
|
| 553 |
+
"eval_wer": 7.321293872298887,
|
| 554 |
+
"step": 700
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 33.02,
|
| 558 |
+
"learning_rate": 1.41e-06,
|
| 559 |
+
"loss": 0.0026,
|
| 560 |
+
"step": 710
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 34.01,
|
| 564 |
+
"learning_rate": 1.26e-06,
|
| 565 |
+
"loss": 0.0024,
|
| 566 |
+
"step": 720
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 34.02,
|
| 570 |
+
"learning_rate": 1.11e-06,
|
| 571 |
+
"loss": 0.0024,
|
| 572 |
+
"step": 730
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 35.01,
|
| 576 |
+
"learning_rate": 9.600000000000001e-07,
|
| 577 |
+
"loss": 0.0026,
|
| 578 |
+
"step": 740
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 35.02,
|
| 582 |
+
"learning_rate": 8.100000000000001e-07,
|
| 583 |
+
"loss": 0.0022,
|
| 584 |
+
"step": 750
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 35.02,
|
| 588 |
+
"eval_loss": 0.399670273065567,
|
| 589 |
+
"eval_runtime": 119.4156,
|
| 590 |
+
"eval_samples_per_second": 2.295,
|
| 591 |
+
"eval_steps_per_second": 0.151,
|
| 592 |
+
"eval_wer": 7.352353906908639,
|
| 593 |
+
"step": 750
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 36.01,
|
| 597 |
+
"learning_rate": 6.6e-07,
|
| 598 |
+
"loss": 0.0023,
|
| 599 |
+
"step": 760
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 36.02,
|
| 603 |
+
"learning_rate": 5.100000000000001e-07,
|
| 604 |
+
"loss": 0.0021,
|
| 605 |
+
"step": 770
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 37.0,
|
| 609 |
+
"learning_rate": 3.6e-07,
|
| 610 |
+
"loss": 0.0022,
|
| 611 |
+
"step": 780
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 37.02,
|
| 615 |
+
"learning_rate": 2.1000000000000003e-07,
|
| 616 |
+
"loss": 0.0021,
|
| 617 |
+
"step": 790
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 38.0,
|
| 621 |
+
"learning_rate": 6.000000000000001e-08,
|
| 622 |
+
"loss": 0.0022,
|
| 623 |
+
"step": 800
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 38.0,
|
| 627 |
+
"eval_loss": 0.4024691879749298,
|
| 628 |
+
"eval_runtime": 131.5096,
|
| 629 |
+
"eval_samples_per_second": 2.083,
|
| 630 |
+
"eval_steps_per_second": 0.137,
|
| 631 |
+
"eval_wer": 7.330168167901673,
|
| 632 |
+
"step": 800
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 38.0,
|
| 636 |
+
"step": 800,
|
| 637 |
+
"total_flos": 2.570087278116864e+19,
|
| 638 |
+
"train_loss": 0.18918936033500358,
|
| 639 |
+
"train_runtime": 13916.8243,
|
| 640 |
+
"train_samples_per_second": 1.84,
|
| 641 |
+
"train_steps_per_second": 0.057
|
| 642 |
+
}
|
| 643 |
+
],
|
| 644 |
+
"max_steps": 800,
|
| 645 |
+
"num_train_epochs": 9223372036854775807,
|
| 646 |
+
"total_flos": 2.570087278116864e+19,
|
| 647 |
+
"trial_name": null,
|
| 648 |
+
"trial_params": null
|
| 649 |
+
}
|