Upload 9 files
Browse files- config.json +1 -1
- optimizer.pt +1 -1
- pytorch_model.bin +1 -1
- rng_state.pth +2 -2
- scaler.pt +1 -1
- scheduler.pt +1 -1
- trainer_state.json +797 -122
- training_args.bin +1 -1
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "
|
| 3 |
"activation_dropout": 0.1,
|
| 4 |
"adapter_kernel_size": 3,
|
| 5 |
"adapter_stride": 2,
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "othrif",
|
| 3 |
"activation_dropout": 0.1,
|
| 4 |
"adapter_kernel_size": 3,
|
| 5 |
"adapter_stride": 2,
|
optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2490594117
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4c79e728d98168d85015d00930e5f1bd407f25c2d89b2d55d9c9bb2f99ce3eee
|
| 3 |
size 2490594117
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262195949
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ccfbee3b3ae99465c34b8091981ba73de06eaab423aefbdb55c7872677393c70
|
| 3 |
size 1262195949
|
rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ef516f13eae1058f7d3a5544d2d46b334d6f3f2c0af866334d159bdf2bf78524
|
| 3 |
+
size 14639
|
scaler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 557
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1cc746e056a378283285a9fcc1e3f23a267ec6b0193f2c4ba34347b78ae0c98f
|
| 3 |
size 557
|
scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6bf093631f76407e20836696a637f5ab6e0d337c65ba9151883e190682646544
|
| 3 |
size 627
|
trainer_state.json
CHANGED
|
@@ -1,211 +1,886 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
-
"epoch":
|
| 12 |
-
"learning_rate": 0.
|
| 13 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
"step": 400
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 2.
|
| 18 |
-
"eval_loss": 0.
|
| 19 |
-
"eval_runtime":
|
| 20 |
-
"eval_samples_per_second":
|
| 21 |
-
"eval_steps_per_second": 2.
|
| 22 |
-
"eval_wer": 0.
|
| 23 |
"step": 400
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch":
|
| 27 |
-
"learning_rate":
|
| 28 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
"step": 800
|
| 30 |
},
|
| 31 |
{
|
| 32 |
-
"epoch":
|
| 33 |
-
"eval_loss": 0.
|
| 34 |
-
"eval_runtime":
|
| 35 |
-
"eval_samples_per_second":
|
| 36 |
-
"eval_steps_per_second": 2.
|
| 37 |
-
"eval_wer": 0.
|
| 38 |
"step": 800
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
"step": 1200
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch":
|
| 48 |
-
"eval_loss": 0.
|
| 49 |
-
"eval_runtime":
|
| 50 |
-
"eval_samples_per_second":
|
| 51 |
-
"eval_steps_per_second": 2.
|
| 52 |
-
"eval_wer": 0.
|
| 53 |
"step": 1200
|
| 54 |
},
|
| 55 |
{
|
| 56 |
-
"epoch": 8.
|
| 57 |
-
"learning_rate":
|
| 58 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
"step": 1600
|
| 60 |
},
|
| 61 |
{
|
| 62 |
-
"epoch":
|
| 63 |
-
"eval_loss": 0.
|
| 64 |
-
"eval_runtime":
|
| 65 |
-
"eval_samples_per_second":
|
| 66 |
-
"eval_steps_per_second": 2.
|
| 67 |
-
"eval_wer": 0.
|
| 68 |
"step": 1600
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch":
|
| 72 |
-
"learning_rate":
|
| 73 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 74 |
"step": 2000
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch":
|
| 78 |
-
"eval_loss": 0.
|
| 79 |
-
"eval_runtime":
|
| 80 |
-
"eval_samples_per_second":
|
| 81 |
-
"eval_steps_per_second": 2.
|
| 82 |
-
"eval_wer": 0.
|
| 83 |
"step": 2000
|
| 84 |
},
|
| 85 |
{
|
| 86 |
-
"epoch":
|
| 87 |
-
"learning_rate":
|
| 88 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 89 |
"step": 2400
|
| 90 |
},
|
| 91 |
{
|
| 92 |
-
"epoch":
|
| 93 |
-
"eval_loss": 0.
|
| 94 |
-
"eval_runtime":
|
| 95 |
-
"eval_samples_per_second":
|
| 96 |
-
"eval_steps_per_second": 2.
|
| 97 |
-
"eval_wer": 0.
|
| 98 |
"step": 2400
|
| 99 |
},
|
| 100 |
{
|
| 101 |
-
"epoch":
|
| 102 |
-
"learning_rate":
|
| 103 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 104 |
"step": 2800
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch":
|
| 108 |
-
"eval_loss": 0.
|
| 109 |
-
"eval_runtime":
|
| 110 |
-
"eval_samples_per_second":
|
| 111 |
-
"eval_steps_per_second":
|
| 112 |
-
"eval_wer": 0.
|
| 113 |
"step": 2800
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"epoch":
|
| 117 |
-
"learning_rate":
|
| 118 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 119 |
"step": 3200
|
| 120 |
},
|
| 121 |
{
|
| 122 |
-
"epoch":
|
| 123 |
-
"eval_loss": 0.
|
| 124 |
-
"eval_runtime":
|
| 125 |
-
"eval_samples_per_second":
|
| 126 |
-
"eval_steps_per_second": 2.
|
| 127 |
-
"eval_wer": 0.
|
| 128 |
"step": 3200
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch":
|
| 132 |
-
"learning_rate":
|
| 133 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 134 |
"step": 3600
|
| 135 |
},
|
| 136 |
{
|
| 137 |
-
"epoch":
|
| 138 |
-
"eval_loss": 0.
|
| 139 |
-
"eval_runtime":
|
| 140 |
-
"eval_samples_per_second":
|
| 141 |
-
"eval_steps_per_second": 2.
|
| 142 |
-
"eval_wer": 0.
|
| 143 |
"step": 3600
|
| 144 |
},
|
| 145 |
{
|
| 146 |
-
"epoch":
|
| 147 |
-
"learning_rate": 7.
|
| 148 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 149 |
"step": 4000
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch":
|
| 153 |
-
"eval_loss": 0.
|
| 154 |
-
"eval_runtime":
|
| 155 |
-
"eval_samples_per_second":
|
| 156 |
-
"eval_steps_per_second": 2.
|
| 157 |
-
"eval_wer": 0.
|
| 158 |
"step": 4000
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch":
|
| 162 |
-
"learning_rate":
|
| 163 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 164 |
"step": 4400
|
| 165 |
},
|
| 166 |
{
|
| 167 |
-
"epoch":
|
| 168 |
-
"eval_loss": 0.
|
| 169 |
-
"eval_runtime":
|
| 170 |
-
"eval_samples_per_second":
|
| 171 |
-
"eval_steps_per_second": 2.
|
| 172 |
-
"eval_wer": 0.
|
| 173 |
"step": 4400
|
| 174 |
},
|
| 175 |
{
|
| 176 |
-
"epoch":
|
| 177 |
-
"learning_rate":
|
| 178 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 179 |
"step": 4800
|
| 180 |
},
|
| 181 |
{
|
| 182 |
-
"epoch":
|
| 183 |
-
"eval_loss": 0.
|
| 184 |
-
"eval_runtime":
|
| 185 |
-
"eval_samples_per_second":
|
| 186 |
-
"eval_steps_per_second": 2.
|
| 187 |
-
"eval_wer": 0.
|
| 188 |
"step": 4800
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch":
|
| 192 |
-
"learning_rate":
|
| 193 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 194 |
"step": 5200
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch":
|
| 198 |
-
"eval_loss": 0.
|
| 199 |
-
"eval_runtime":
|
| 200 |
-
"eval_samples_per_second":
|
| 201 |
-
"eval_steps_per_second": 2.
|
| 202 |
-
"eval_wer": 0.
|
| 203 |
"step": 5200
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 204 |
}
|
| 205 |
],
|
| 206 |
-
"max_steps":
|
| 207 |
-
"num_train_epochs":
|
| 208 |
-
"total_flos":
|
| 209 |
"trial_name": null,
|
| 210 |
"trial_params": null
|
| 211 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 37.90553745928339,
|
| 5 |
+
"global_step": 5800,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.65,
|
| 12 |
+
"learning_rate": 0.0001,
|
| 13 |
+
"loss": 1.0071,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.65,
|
| 18 |
+
"eval_loss": 0.3728577196598053,
|
| 19 |
+
"eval_runtime": 169.6136,
|
| 20 |
+
"eval_samples_per_second": 19.48,
|
| 21 |
+
"eval_steps_per_second": 2.435,
|
| 22 |
+
"eval_wer": 0.5494842925753878,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.31,
|
| 27 |
+
"learning_rate": 9.918652891889694e-05,
|
| 28 |
+
"loss": 0.6823,
|
| 29 |
+
"step": 200
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 1.31,
|
| 33 |
+
"eval_loss": 0.33882805705070496,
|
| 34 |
+
"eval_runtime": 167.327,
|
| 35 |
+
"eval_samples_per_second": 19.746,
|
| 36 |
+
"eval_steps_per_second": 2.468,
|
| 37 |
+
"eval_wer": 0.5344461066057791,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.96,
|
| 42 |
+
"learning_rate": 9.837305783779387e-05,
|
| 43 |
+
"loss": 0.6063,
|
| 44 |
+
"step": 300
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.96,
|
| 48 |
+
"eval_loss": 0.3208909332752228,
|
| 49 |
+
"eval_runtime": 174.661,
|
| 50 |
+
"eval_samples_per_second": 18.917,
|
| 51 |
+
"eval_steps_per_second": 2.365,
|
| 52 |
+
"eval_wer": 0.5098811117234864,
|
| 53 |
+
"step": 300
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 2.61,
|
| 57 |
+
"learning_rate": 9.755958675669081e-05,
|
| 58 |
+
"loss": 0.5326,
|
| 59 |
"step": 400
|
| 60 |
},
|
| 61 |
{
|
| 62 |
+
"epoch": 2.61,
|
| 63 |
+
"eval_loss": 0.30963289737701416,
|
| 64 |
+
"eval_runtime": 177.329,
|
| 65 |
+
"eval_samples_per_second": 18.632,
|
| 66 |
+
"eval_steps_per_second": 2.329,
|
| 67 |
+
"eval_wer": 0.5025588536335721,
|
| 68 |
"step": 400
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 3.27,
|
| 72 |
+
"learning_rate": 9.674611567558773e-05,
|
| 73 |
+
"loss": 0.5074,
|
| 74 |
+
"step": 500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 3.27,
|
| 78 |
+
"eval_loss": 0.3118290901184082,
|
| 79 |
+
"eval_runtime": 173.1863,
|
| 80 |
+
"eval_samples_per_second": 19.078,
|
| 81 |
+
"eval_steps_per_second": 2.385,
|
| 82 |
+
"eval_wer": 0.4959452011652626,
|
| 83 |
+
"step": 500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 3.92,
|
| 87 |
+
"learning_rate": 9.593264459448467e-05,
|
| 88 |
+
"loss": 0.446,
|
| 89 |
+
"step": 600
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 3.92,
|
| 93 |
+
"eval_loss": 0.30445897579193115,
|
| 94 |
+
"eval_runtime": 173.9457,
|
| 95 |
+
"eval_samples_per_second": 18.994,
|
| 96 |
+
"eval_steps_per_second": 2.374,
|
| 97 |
+
"eval_wer": 0.4865758601684907,
|
| 98 |
+
"step": 600
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 4.57,
|
| 102 |
+
"learning_rate": 9.51191735133816e-05,
|
| 103 |
+
"loss": 0.4283,
|
| 104 |
+
"step": 700
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 4.57,
|
| 108 |
+
"eval_loss": 0.30916285514831543,
|
| 109 |
+
"eval_runtime": 176.2446,
|
| 110 |
+
"eval_samples_per_second": 18.747,
|
| 111 |
+
"eval_steps_per_second": 2.343,
|
| 112 |
+
"eval_wer": 0.48366270372411624,
|
| 113 |
+
"step": 700
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 5.23,
|
| 117 |
+
"learning_rate": 9.430570243227855e-05,
|
| 118 |
+
"loss": 0.41,
|
| 119 |
"step": 800
|
| 120 |
},
|
| 121 |
{
|
| 122 |
+
"epoch": 5.23,
|
| 123 |
+
"eval_loss": 0.35956883430480957,
|
| 124 |
+
"eval_runtime": 169.7235,
|
| 125 |
+
"eval_samples_per_second": 19.467,
|
| 126 |
+
"eval_steps_per_second": 2.433,
|
| 127 |
+
"eval_wer": 0.4932682465947563,
|
| 128 |
"step": 800
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 5.88,
|
| 132 |
+
"learning_rate": 9.349223135117547e-05,
|
| 133 |
+
"loss": 0.3802,
|
| 134 |
+
"step": 900
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 5.88,
|
| 138 |
+
"eval_loss": 0.3234783411026001,
|
| 139 |
+
"eval_runtime": 171.4522,
|
| 140 |
+
"eval_samples_per_second": 19.271,
|
| 141 |
+
"eval_steps_per_second": 2.409,
|
| 142 |
+
"eval_wer": 0.47752145500354304,
|
| 143 |
+
"step": 900
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 6.53,
|
| 147 |
+
"learning_rate": 9.267876027007241e-05,
|
| 148 |
+
"loss": 0.3852,
|
| 149 |
+
"step": 1000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 6.53,
|
| 153 |
+
"eval_loss": 0.32342973351478577,
|
| 154 |
+
"eval_runtime": 184.4459,
|
| 155 |
+
"eval_samples_per_second": 17.913,
|
| 156 |
+
"eval_steps_per_second": 2.239,
|
| 157 |
+
"eval_wer": 0.47815132666719157,
|
| 158 |
+
"step": 1000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 7.19,
|
| 162 |
+
"learning_rate": 9.186528918896934e-05,
|
| 163 |
+
"loss": 0.3539,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 7.19,
|
| 168 |
+
"eval_loss": 0.33684083819389343,
|
| 169 |
+
"eval_runtime": 170.8359,
|
| 170 |
+
"eval_samples_per_second": 19.34,
|
| 171 |
+
"eval_steps_per_second": 2.418,
|
| 172 |
+
"eval_wer": 0.4796472718683568,
|
| 173 |
+
"step": 1100
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 7.84,
|
| 177 |
+
"learning_rate": 9.105181810786628e-05,
|
| 178 |
+
"loss": 0.3444,
|
| 179 |
"step": 1200
|
| 180 |
},
|
| 181 |
{
|
| 182 |
+
"epoch": 7.84,
|
| 183 |
+
"eval_loss": 0.3268304169178009,
|
| 184 |
+
"eval_runtime": 170.984,
|
| 185 |
+
"eval_samples_per_second": 19.323,
|
| 186 |
+
"eval_steps_per_second": 2.415,
|
| 187 |
+
"eval_wer": 0.4732698212739154,
|
| 188 |
"step": 1200
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 8.5,
|
| 192 |
+
"learning_rate": 9.02383470267632e-05,
|
| 193 |
+
"loss": 0.336,
|
| 194 |
+
"step": 1300
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 8.5,
|
| 198 |
+
"eval_loss": 0.34285250306129456,
|
| 199 |
+
"eval_runtime": 171.7981,
|
| 200 |
+
"eval_samples_per_second": 19.232,
|
| 201 |
+
"eval_steps_per_second": 2.404,
|
| 202 |
+
"eval_wer": 0.479883473742225,
|
| 203 |
+
"step": 1300
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 9.15,
|
| 207 |
+
"learning_rate": 8.942487594566014e-05,
|
| 208 |
+
"loss": 0.3041,
|
| 209 |
+
"step": 1400
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 9.15,
|
| 213 |
+
"eval_loss": 0.35453349351882935,
|
| 214 |
+
"eval_runtime": 172.0678,
|
| 215 |
+
"eval_samples_per_second": 19.202,
|
| 216 |
+
"eval_steps_per_second": 2.4,
|
| 217 |
+
"eval_wer": 0.46248326903393433,
|
| 218 |
+
"step": 1400
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 9.8,
|
| 222 |
+
"learning_rate": 8.861140486455706e-05,
|
| 223 |
+
"loss": 0.3074,
|
| 224 |
+
"step": 1500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 9.8,
|
| 228 |
+
"eval_loss": 0.3339354693889618,
|
| 229 |
+
"eval_runtime": 172.1803,
|
| 230 |
+
"eval_samples_per_second": 19.189,
|
| 231 |
+
"eval_steps_per_second": 2.399,
|
| 232 |
+
"eval_wer": 0.46319187465553896,
|
| 233 |
+
"step": 1500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 10.46,
|
| 237 |
+
"learning_rate": 8.7797933783454e-05,
|
| 238 |
+
"loss": 0.2948,
|
| 239 |
"step": 1600
|
| 240 |
},
|
| 241 |
{
|
| 242 |
+
"epoch": 10.46,
|
| 243 |
+
"eval_loss": 0.34325212240219116,
|
| 244 |
+
"eval_runtime": 171.9876,
|
| 245 |
+
"eval_samples_per_second": 19.211,
|
| 246 |
+
"eval_steps_per_second": 2.401,
|
| 247 |
+
"eval_wer": 0.4646090858987481,
|
| 248 |
"step": 1600
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 11.11,
|
| 252 |
+
"learning_rate": 8.698446270235093e-05,
|
| 253 |
+
"loss": 0.2905,
|
| 254 |
+
"step": 1700
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 11.11,
|
| 258 |
+
"eval_loss": 0.34282687306404114,
|
| 259 |
+
"eval_runtime": 172.5719,
|
| 260 |
+
"eval_samples_per_second": 19.146,
|
| 261 |
+
"eval_steps_per_second": 2.393,
|
| 262 |
+
"eval_wer": 0.4641366821510117,
|
| 263 |
+
"step": 1700
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 11.76,
|
| 267 |
+
"learning_rate": 8.617099162124787e-05,
|
| 268 |
+
"loss": 0.296,
|
| 269 |
+
"step": 1800
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 11.76,
|
| 273 |
+
"eval_loss": 0.35734105110168457,
|
| 274 |
+
"eval_runtime": 173.1126,
|
| 275 |
+
"eval_samples_per_second": 19.086,
|
| 276 |
+
"eval_steps_per_second": 2.386,
|
| 277 |
+
"eval_wer": 0.4665774348476498,
|
| 278 |
+
"step": 1800
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 12.42,
|
| 282 |
+
"learning_rate": 8.535752054014479e-05,
|
| 283 |
+
"loss": 0.2669,
|
| 284 |
+
"step": 1900
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 12.42,
|
| 288 |
+
"eval_loss": 0.34095147252082825,
|
| 289 |
+
"eval_runtime": 172.9162,
|
| 290 |
+
"eval_samples_per_second": 19.108,
|
| 291 |
+
"eval_steps_per_second": 2.388,
|
| 292 |
+
"eval_wer": 0.46224706716006614,
|
| 293 |
+
"step": 1900
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 13.07,
|
| 297 |
+
"learning_rate": 8.454404945904173e-05,
|
| 298 |
+
"loss": 0.2778,
|
| 299 |
"step": 2000
|
| 300 |
},
|
| 301 |
{
|
| 302 |
+
"epoch": 13.07,
|
| 303 |
+
"eval_loss": 0.3445983827114105,
|
| 304 |
+
"eval_runtime": 172.7429,
|
| 305 |
+
"eval_samples_per_second": 19.127,
|
| 306 |
+
"eval_steps_per_second": 2.391,
|
| 307 |
+
"eval_wer": 0.4621683332021101,
|
| 308 |
"step": 2000
|
| 309 |
},
|
| 310 |
{
|
| 311 |
+
"epoch": 13.72,
|
| 312 |
+
"learning_rate": 8.373057837793867e-05,
|
| 313 |
+
"loss": 0.2605,
|
| 314 |
+
"step": 2100
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 13.72,
|
| 318 |
+
"eval_loss": 0.364580363035202,
|
| 319 |
+
"eval_runtime": 173.3351,
|
| 320 |
+
"eval_samples_per_second": 19.061,
|
| 321 |
+
"eval_steps_per_second": 2.383,
|
| 322 |
+
"eval_wer": 0.4611447917486812,
|
| 323 |
+
"step": 2100
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 14.38,
|
| 327 |
+
"learning_rate": 8.291710729683561e-05,
|
| 328 |
+
"loss": 0.2562,
|
| 329 |
+
"step": 2200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 14.38,
|
| 333 |
+
"eval_loss": 0.3529307544231415,
|
| 334 |
+
"eval_runtime": 173.4538,
|
| 335 |
+
"eval_samples_per_second": 19.048,
|
| 336 |
+
"eval_steps_per_second": 2.381,
|
| 337 |
+
"eval_wer": 0.46201086528619795,
|
| 338 |
+
"step": 2200
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 15.03,
|
| 342 |
+
"learning_rate": 8.210363621573253e-05,
|
| 343 |
+
"loss": 0.2587,
|
| 344 |
+
"step": 2300
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 15.03,
|
| 348 |
+
"eval_loss": 0.35722818970680237,
|
| 349 |
+
"eval_runtime": 173.1723,
|
| 350 |
+
"eval_samples_per_second": 19.079,
|
| 351 |
+
"eval_steps_per_second": 2.385,
|
| 352 |
+
"eval_wer": 0.4694118573340682,
|
| 353 |
+
"step": 2300
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 15.68,
|
| 357 |
+
"learning_rate": 8.129016513462947e-05,
|
| 358 |
+
"loss": 0.242,
|
| 359 |
"step": 2400
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 15.68,
|
| 363 |
+
"eval_loss": 0.36534029245376587,
|
| 364 |
+
"eval_runtime": 173.1065,
|
| 365 |
+
"eval_samples_per_second": 19.087,
|
| 366 |
+
"eval_steps_per_second": 2.386,
|
| 367 |
+
"eval_wer": 0.45894024092591135,
|
| 368 |
"step": 2400
|
| 369 |
},
|
| 370 |
{
|
| 371 |
+
"epoch": 16.34,
|
| 372 |
+
"learning_rate": 8.047669405352641e-05,
|
| 373 |
+
"loss": 0.232,
|
| 374 |
+
"step": 2500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 16.34,
|
| 378 |
+
"eval_loss": 0.34964719414711,
|
| 379 |
+
"eval_runtime": 174.2382,
|
| 380 |
+
"eval_samples_per_second": 18.963,
|
| 381 |
+
"eval_steps_per_second": 2.37,
|
| 382 |
+
"eval_wer": 0.4605149200850327,
|
| 383 |
+
"step": 2500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 16.99,
|
| 387 |
+
"learning_rate": 7.966322297242333e-05,
|
| 388 |
+
"loss": 0.2474,
|
| 389 |
+
"step": 2600
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 16.99,
|
| 393 |
+
"eval_loss": 0.3596344590187073,
|
| 394 |
+
"eval_runtime": 174.0298,
|
| 395 |
+
"eval_samples_per_second": 18.985,
|
| 396 |
+
"eval_steps_per_second": 2.373,
|
| 397 |
+
"eval_wer": 0.46783717817494685,
|
| 398 |
+
"step": 2600
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 17.64,
|
| 402 |
+
"learning_rate": 7.884975189132027e-05,
|
| 403 |
+
"loss": 0.2137,
|
| 404 |
+
"step": 2700
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 17.64,
|
| 408 |
+
"eval_loss": 0.3547351360321045,
|
| 409 |
+
"eval_runtime": 174.6108,
|
| 410 |
+
"eval_samples_per_second": 18.922,
|
| 411 |
+
"eval_steps_per_second": 2.365,
|
| 412 |
+
"eval_wer": 0.4609873238327691,
|
| 413 |
+
"step": 2700
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 18.3,
|
| 417 |
+
"learning_rate": 7.80362808102172e-05,
|
| 418 |
+
"loss": 0.2261,
|
| 419 |
"step": 2800
|
| 420 |
},
|
| 421 |
{
|
| 422 |
+
"epoch": 18.3,
|
| 423 |
+
"eval_loss": 0.35713937878608704,
|
| 424 |
+
"eval_runtime": 173.8691,
|
| 425 |
+
"eval_samples_per_second": 19.003,
|
| 426 |
+
"eval_steps_per_second": 2.375,
|
| 427 |
+
"eval_wer": 0.4579954334304385,
|
| 428 |
"step": 2800
|
| 429 |
},
|
| 430 |
{
|
| 431 |
+
"epoch": 18.95,
|
| 432 |
+
"learning_rate": 7.723094443992517e-05,
|
| 433 |
+
"loss": 0.2141,
|
| 434 |
+
"step": 2900
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 18.95,
|
| 438 |
+
"eval_loss": 0.36411064863204956,
|
| 439 |
+
"eval_runtime": 174.3463,
|
| 440 |
+
"eval_samples_per_second": 18.951,
|
| 441 |
+
"eval_steps_per_second": 2.369,
|
| 442 |
+
"eval_wer": 0.45563341469175656,
|
| 443 |
+
"step": 2900
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 19.61,
|
| 447 |
+
"learning_rate": 7.64174733588221e-05,
|
| 448 |
+
"loss": 0.2201,
|
| 449 |
+
"step": 3000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 19.61,
|
| 453 |
+
"eval_loss": 0.34566032886505127,
|
| 454 |
+
"eval_runtime": 173.9331,
|
| 455 |
+
"eval_samples_per_second": 18.996,
|
| 456 |
+
"eval_steps_per_second": 2.374,
|
| 457 |
+
"eval_wer": 0.45303519407920634,
|
| 458 |
+
"step": 3000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 20.26,
|
| 462 |
+
"learning_rate": 7.560400227771903e-05,
|
| 463 |
+
"loss": 0.2243,
|
| 464 |
+
"step": 3100
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 20.26,
|
| 468 |
+
"eval_loss": 0.3523178994655609,
|
| 469 |
+
"eval_runtime": 174.1671,
|
| 470 |
+
"eval_samples_per_second": 18.97,
|
| 471 |
+
"eval_steps_per_second": 2.371,
|
| 472 |
+
"eval_wer": 0.4571293598929218,
|
| 473 |
+
"step": 3100
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 20.91,
|
| 477 |
+
"learning_rate": 7.479053119661597e-05,
|
| 478 |
+
"loss": 0.1891,
|
| 479 |
"step": 3200
|
| 480 |
},
|
| 481 |
{
|
| 482 |
+
"epoch": 20.91,
|
| 483 |
+
"eval_loss": 0.337533563375473,
|
| 484 |
+
"eval_runtime": 174.612,
|
| 485 |
+
"eval_samples_per_second": 18.922,
|
| 486 |
+
"eval_steps_per_second": 2.365,
|
| 487 |
+
"eval_wer": 0.4541374694905913,
|
| 488 |
"step": 3200
|
| 489 |
},
|
| 490 |
{
|
| 491 |
+
"epoch": 21.57,
|
| 492 |
+
"learning_rate": 7.39770601155129e-05,
|
| 493 |
+
"loss": 0.2033,
|
| 494 |
+
"step": 3300
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 21.57,
|
| 498 |
+
"eval_loss": 0.3634466230869293,
|
| 499 |
+
"eval_runtime": 174.6521,
|
| 500 |
+
"eval_samples_per_second": 18.918,
|
| 501 |
+
"eval_steps_per_second": 2.365,
|
| 502 |
+
"eval_wer": 0.4579166994724825,
|
| 503 |
+
"step": 3300
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 22.22,
|
| 507 |
+
"learning_rate": 7.316358903440983e-05,
|
| 508 |
+
"loss": 0.2035,
|
| 509 |
+
"step": 3400
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 22.22,
|
| 513 |
+
"eval_loss": 0.3793589174747467,
|
| 514 |
+
"eval_runtime": 174.394,
|
| 515 |
+
"eval_samples_per_second": 18.946,
|
| 516 |
+
"eval_steps_per_second": 2.368,
|
| 517 |
+
"eval_wer": 0.4555546807338005,
|
| 518 |
+
"step": 3400
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 22.87,
|
| 522 |
+
"learning_rate": 7.235011795330676e-05,
|
| 523 |
+
"loss": 0.1867,
|
| 524 |
+
"step": 3500
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 22.87,
|
| 528 |
+
"eval_loss": 0.37910905480384827,
|
| 529 |
+
"eval_runtime": 174.9971,
|
| 530 |
+
"eval_samples_per_second": 18.88,
|
| 531 |
+
"eval_steps_per_second": 2.36,
|
| 532 |
+
"eval_wer": 0.454924809070152,
|
| 533 |
+
"step": 3500
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 23.53,
|
| 537 |
+
"learning_rate": 7.15366468722037e-05,
|
| 538 |
+
"loss": 0.1956,
|
| 539 |
"step": 3600
|
| 540 |
},
|
| 541 |
{
|
| 542 |
+
"epoch": 23.53,
|
| 543 |
+
"eval_loss": 0.3568515479564667,
|
| 544 |
+
"eval_runtime": 174.799,
|
| 545 |
+
"eval_samples_per_second": 18.902,
|
| 546 |
+
"eval_steps_per_second": 2.363,
|
| 547 |
+
"eval_wer": 0.45760176364065824,
|
| 548 |
"step": 3600
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 24.18,
|
| 552 |
+
"learning_rate": 7.072317579110062e-05,
|
| 553 |
+
"loss": 0.1826,
|
| 554 |
+
"step": 3700
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 24.18,
|
| 558 |
+
"eval_loss": 0.3747410178184509,
|
| 559 |
+
"eval_runtime": 175.1918,
|
| 560 |
+
"eval_samples_per_second": 18.859,
|
| 561 |
+
"eval_steps_per_second": 2.357,
|
| 562 |
+
"eval_wer": 0.4543736713644595,
|
| 563 |
+
"step": 3700
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 24.83,
|
| 567 |
+
"learning_rate": 6.99178394208086e-05,
|
| 568 |
+
"loss": 0.1867,
|
| 569 |
+
"step": 3800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 24.83,
|
| 573 |
+
"eval_loss": 0.36731651425361633,
|
| 574 |
+
"eval_runtime": 175.3726,
|
| 575 |
+
"eval_samples_per_second": 18.84,
|
| 576 |
+
"eval_steps_per_second": 2.355,
|
| 577 |
+
"eval_wer": 0.45366506574285487,
|
| 578 |
+
"step": 3800
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 25.49,
|
| 582 |
+
"learning_rate": 6.910436833970553e-05,
|
| 583 |
+
"loss": 0.1902,
|
| 584 |
+
"step": 3900
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 25.49,
|
| 588 |
+
"eval_loss": 0.3835786283016205,
|
| 589 |
+
"eval_runtime": 182.8434,
|
| 590 |
+
"eval_samples_per_second": 18.07,
|
| 591 |
+
"eval_steps_per_second": 2.259,
|
| 592 |
+
"eval_wer": 0.4522478544996457,
|
| 593 |
+
"step": 3900
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 26.14,
|
| 597 |
+
"learning_rate": 6.829089725860246e-05,
|
| 598 |
+
"loss": 0.1786,
|
| 599 |
"step": 4000
|
| 600 |
},
|
| 601 |
{
|
| 602 |
+
"epoch": 26.14,
|
| 603 |
+
"eval_loss": 0.3528241813182831,
|
| 604 |
+
"eval_runtime": 182.8588,
|
| 605 |
+
"eval_samples_per_second": 18.069,
|
| 606 |
+
"eval_steps_per_second": 2.259,
|
| 607 |
+
"eval_wer": 0.4485473584757106,
|
| 608 |
"step": 4000
|
| 609 |
},
|
| 610 |
{
|
| 611 |
+
"epoch": 26.79,
|
| 612 |
+
"learning_rate": 6.74774261774994e-05,
|
| 613 |
+
"loss": 0.178,
|
| 614 |
+
"step": 4100
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 26.79,
|
| 618 |
+
"eval_loss": 0.3756342828273773,
|
| 619 |
+
"eval_runtime": 183.4843,
|
| 620 |
+
"eval_samples_per_second": 18.007,
|
| 621 |
+
"eval_steps_per_second": 2.251,
|
| 622 |
+
"eval_wer": 0.45303519407920634,
|
| 623 |
+
"step": 4100
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 27.45,
|
| 627 |
+
"learning_rate": 6.666395509639632e-05,
|
| 628 |
+
"loss": 0.1783,
|
| 629 |
+
"step": 4200
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 27.45,
|
| 633 |
+
"eval_loss": 0.38552403450012207,
|
| 634 |
+
"eval_runtime": 185.3761,
|
| 635 |
+
"eval_samples_per_second": 17.823,
|
| 636 |
+
"eval_steps_per_second": 2.228,
|
| 637 |
+
"eval_wer": 0.4515392488780411,
|
| 638 |
+
"step": 4200
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 28.1,
|
| 642 |
+
"learning_rate": 6.585048401529326e-05,
|
| 643 |
+
"loss": 0.1747,
|
| 644 |
+
"step": 4300
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 28.1,
|
| 648 |
+
"eval_loss": 0.3594723045825958,
|
| 649 |
+
"eval_runtime": 184.1815,
|
| 650 |
+
"eval_samples_per_second": 17.939,
|
| 651 |
+
"eval_steps_per_second": 2.242,
|
| 652 |
+
"eval_wer": 0.4475238170222817,
|
| 653 |
+
"step": 4300
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 28.76,
|
| 657 |
+
"learning_rate": 6.503701293419018e-05,
|
| 658 |
+
"loss": 0.1776,
|
| 659 |
"step": 4400
|
| 660 |
},
|
| 661 |
{
|
| 662 |
+
"epoch": 28.76,
|
| 663 |
+
"eval_loss": 0.3899536728858948,
|
| 664 |
+
"eval_runtime": 183.8028,
|
| 665 |
+
"eval_samples_per_second": 17.976,
|
| 666 |
+
"eval_steps_per_second": 2.247,
|
| 667 |
+
"eval_wer": 0.45303519407920634,
|
| 668 |
"step": 4400
|
| 669 |
},
|
| 670 |
{
|
| 671 |
+
"epoch": 29.41,
|
| 672 |
+
"learning_rate": 6.422354185308712e-05,
|
| 673 |
+
"loss": 0.1615,
|
| 674 |
+
"step": 4500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 29.41,
|
| 678 |
+
"eval_loss": 0.37925612926483154,
|
| 679 |
+
"eval_runtime": 184.3645,
|
| 680 |
+
"eval_samples_per_second": 17.921,
|
| 681 |
+
"eval_steps_per_second": 2.24,
|
| 682 |
+
"eval_wer": 0.4487048263916227,
|
| 683 |
+
"step": 4500
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 30.07,
|
| 687 |
+
"learning_rate": 6.341007077198405e-05,
|
| 688 |
+
"loss": 0.1665,
|
| 689 |
+
"step": 4600
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 30.07,
|
| 693 |
+
"eval_loss": 0.3769548237323761,
|
| 694 |
+
"eval_runtime": 185.5661,
|
| 695 |
+
"eval_samples_per_second": 17.805,
|
| 696 |
+
"eval_steps_per_second": 2.226,
|
| 697 |
+
"eval_wer": 0.4504369734666562,
|
| 698 |
+
"step": 4600
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 30.72,
|
| 702 |
+
"learning_rate": 6.2596599690881e-05,
|
| 703 |
+
"loss": 0.1562,
|
| 704 |
+
"step": 4700
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 30.72,
|
| 708 |
+
"eval_loss": 0.38725826144218445,
|
| 709 |
+
"eval_runtime": 184.2843,
|
| 710 |
+
"eval_samples_per_second": 17.929,
|
| 711 |
+
"eval_steps_per_second": 2.241,
|
| 712 |
+
"eval_wer": 0.45090937721439256,
|
| 713 |
+
"step": 4700
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 31.37,
|
| 717 |
+
"learning_rate": 6.178312860977793e-05,
|
| 718 |
+
"loss": 0.1558,
|
| 719 |
"step": 4800
|
| 720 |
},
|
| 721 |
{
|
| 722 |
+
"epoch": 31.37,
|
| 723 |
+
"eval_loss": 0.37403690814971924,
|
| 724 |
+
"eval_runtime": 184.7842,
|
| 725 |
+
"eval_samples_per_second": 17.88,
|
| 726 |
+
"eval_steps_per_second": 2.235,
|
| 727 |
+
"eval_wer": 0.4494134320132273,
|
| 728 |
"step": 4800
|
| 729 |
},
|
| 730 |
{
|
| 731 |
+
"epoch": 32.03,
|
| 732 |
+
"learning_rate": 6.0969657528674864e-05,
|
| 733 |
+
"loss": 0.1574,
|
| 734 |
+
"step": 4900
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 32.03,
|
| 738 |
+
"eval_loss": 0.38782382011413574,
|
| 739 |
+
"eval_runtime": 185.4497,
|
| 740 |
+
"eval_samples_per_second": 17.816,
|
| 741 |
+
"eval_steps_per_second": 2.227,
|
| 742 |
+
"eval_wer": 0.44933469805527126,
|
| 743 |
+
"step": 4900
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 32.68,
|
| 747 |
+
"learning_rate": 6.0156186447571796e-05,
|
| 748 |
+
"loss": 0.152,
|
| 749 |
+
"step": 5000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 32.68,
|
| 753 |
+
"eval_loss": 0.36702463030815125,
|
| 754 |
+
"eval_runtime": 184.067,
|
| 755 |
+
"eval_samples_per_second": 17.95,
|
| 756 |
+
"eval_steps_per_second": 2.244,
|
| 757 |
+
"eval_wer": 0.44933469805527126,
|
| 758 |
+
"step": 5000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 33.33,
|
| 762 |
+
"learning_rate": 5.934271536646873e-05,
|
| 763 |
+
"loss": 0.1477,
|
| 764 |
+
"step": 5100
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 33.33,
|
| 768 |
+
"eval_loss": 0.36524683237075806,
|
| 769 |
+
"eval_runtime": 186.7144,
|
| 770 |
+
"eval_samples_per_second": 17.695,
|
| 771 |
+
"eval_steps_per_second": 2.212,
|
| 772 |
+
"eval_wer": 0.4496496338870955,
|
| 773 |
+
"step": 5100
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 33.98,
|
| 777 |
+
"learning_rate": 5.852924428536566e-05,
|
| 778 |
+
"loss": 0.1561,
|
| 779 |
"step": 5200
|
| 780 |
},
|
| 781 |
{
|
| 782 |
+
"epoch": 33.98,
|
| 783 |
+
"eval_loss": 0.3987789452075958,
|
| 784 |
+
"eval_runtime": 177.1487,
|
| 785 |
+
"eval_samples_per_second": 18.651,
|
| 786 |
+
"eval_steps_per_second": 2.331,
|
| 787 |
+
"eval_wer": 0.4535863317848988,
|
| 788 |
"step": 5200
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 34.64,
|
| 792 |
+
"learning_rate": 5.771577320426259e-05,
|
| 793 |
+
"loss": 0.1441,
|
| 794 |
+
"step": 5300
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 34.64,
|
| 798 |
+
"eval_loss": 0.37290704250335693,
|
| 799 |
+
"eval_runtime": 178.4122,
|
| 800 |
+
"eval_samples_per_second": 18.519,
|
| 801 |
+
"eval_steps_per_second": 2.315,
|
| 802 |
+
"eval_wer": 0.4471301472325014,
|
| 803 |
+
"step": 5300
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 35.29,
|
| 807 |
+
"learning_rate": 5.691043683397056e-05,
|
| 808 |
+
"loss": 0.1462,
|
| 809 |
+
"step": 5400
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 35.29,
|
| 813 |
+
"eval_loss": 0.3913721740245819,
|
| 814 |
+
"eval_runtime": 175.0751,
|
| 815 |
+
"eval_samples_per_second": 18.872,
|
| 816 |
+
"eval_steps_per_second": 2.359,
|
| 817 |
+
"eval_wer": 0.4488622943075348,
|
| 818 |
+
"step": 5400
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 35.94,
|
| 822 |
+
"learning_rate": 5.6096965752867494e-05,
|
| 823 |
+
"loss": 0.1388,
|
| 824 |
+
"step": 5500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 35.94,
|
| 828 |
+
"eval_loss": 0.3886808454990387,
|
| 829 |
+
"eval_runtime": 175.0289,
|
| 830 |
+
"eval_samples_per_second": 18.877,
|
| 831 |
+
"eval_steps_per_second": 2.36,
|
| 832 |
+
"eval_wer": 0.44807495472797415,
|
| 833 |
+
"step": 5500
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 36.6,
|
| 837 |
+
"learning_rate": 5.5283494671764426e-05,
|
| 838 |
+
"loss": 0.1362,
|
| 839 |
+
"step": 5600
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 36.6,
|
| 843 |
+
"eval_loss": 0.3816515803337097,
|
| 844 |
+
"eval_runtime": 175.1136,
|
| 845 |
+
"eval_samples_per_second": 18.868,
|
| 846 |
+
"eval_steps_per_second": 2.358,
|
| 847 |
+
"eval_wer": 0.445476734115424,
|
| 848 |
+
"step": 5600
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 37.25,
|
| 852 |
+
"learning_rate": 5.447002359066136e-05,
|
| 853 |
+
"loss": 0.1439,
|
| 854 |
+
"step": 5700
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 37.25,
|
| 858 |
+
"eval_loss": 0.39244014024734497,
|
| 859 |
+
"eval_runtime": 175.9891,
|
| 860 |
+
"eval_samples_per_second": 18.774,
|
| 861 |
+
"eval_steps_per_second": 2.347,
|
| 862 |
+
"eval_wer": 0.446657743484765,
|
| 863 |
+
"step": 5700
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 37.91,
|
| 867 |
+
"learning_rate": 5.365655250955829e-05,
|
| 868 |
+
"loss": 0.1299,
|
| 869 |
+
"step": 5800
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 37.91,
|
| 873 |
+
"eval_loss": 0.3693729341030121,
|
| 874 |
+
"eval_runtime": 176.2438,
|
| 875 |
+
"eval_samples_per_second": 18.747,
|
| 876 |
+
"eval_steps_per_second": 2.343,
|
| 877 |
+
"eval_wer": 0.4430359814187859,
|
| 878 |
+
"step": 5800
|
| 879 |
}
|
| 880 |
],
|
| 881 |
+
"max_steps": 12393,
|
| 882 |
+
"num_train_epochs": 81,
|
| 883 |
+
"total_flos": 2.5038038866869117e+19,
|
| 884 |
"trial_name": null,
|
| 885 |
"trial_params": null
|
| 886 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3003
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:429d56e7f33237cdebb585dfadbe12372aa7b7c12ffbf8faf5185cef71f533cb
|
| 3 |
size 3003
|