Upload 9 files
Browse files- optimizer.pt +2 -2
- pytorch_model.bin +1 -1
- rng_state.pth +1 -1
- scaler.pt +1 -1
- scheduler.pt +1 -1
- trainer_state.json +387 -3
optimizer.pt
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:cc97aa00e9568bef0834b31d19ece717bef392afdd8bb997f73b330a83cc66f3
|
| 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 1262168365
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cf326143bd61e53e8d2864891d8c4ae23cbcacb66a92a67cee3770bf806840c7
|
| 3 |
size 1262168365
|
rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c8dc1ae25e326d697b6f22afe1e02c2973a5a863e66e3d5357f98ec0113d01c8
|
| 3 |
size 14575
|
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:c28126f0ef915e3d7dfb516546fa859e01feb22702842408747aacd7837fd408
|
| 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 sha256:245bdbe880e4352551249eb1a6870d96668e498aa4cf925daf7f4cc205d39308
|
| 3 |
size 627
|
trainer_state.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 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,
|
|
@@ -27,11 +27,395 @@
|
|
| 27 |
"eval_steps_per_second": 1.699,
|
| 28 |
"eval_wer": 0.18217074305806918,
|
| 29 |
"step": 250
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
}
|
| 31 |
],
|
| 32 |
"max_steps": 1340000,
|
| 33 |
"num_train_epochs": 10000,
|
| 34 |
-
"total_flos":
|
| 35 |
"trial_name": null,
|
| 36 |
"trial_params": null
|
| 37 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 31.71508379888268,
|
| 5 |
+
"global_step": 4250,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 27 |
"eval_steps_per_second": 1.699,
|
| 28 |
"eval_wer": 0.18217074305806918,
|
| 29 |
"step": 250
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 2.24,
|
| 33 |
+
"learning_rate": 4.999253675647437e-05,
|
| 34 |
+
"loss": 0.0381,
|
| 35 |
+
"step": 300
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 2.98,
|
| 39 |
+
"learning_rate": 4.998884245092918e-05,
|
| 40 |
+
"loss": 0.0343,
|
| 41 |
+
"step": 400
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 3.73,
|
| 45 |
+
"learning_rate": 4.998511082916636e-05,
|
| 46 |
+
"loss": 0.0325,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 3.73,
|
| 51 |
+
"eval_loss": 0.2165367156267166,
|
| 52 |
+
"eval_runtime": 199.6867,
|
| 53 |
+
"eval_samples_per_second": 19.195,
|
| 54 |
+
"eval_steps_per_second": 2.404,
|
| 55 |
+
"eval_wer": 0.18279125084027095,
|
| 56 |
+
"step": 500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 4.48,
|
| 60 |
+
"learning_rate": 4.998137920740354e-05,
|
| 61 |
+
"loss": 0.0324,
|
| 62 |
+
"step": 600
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 5.22,
|
| 66 |
+
"learning_rate": 4.9977647585640724e-05,
|
| 67 |
+
"loss": 0.0303,
|
| 68 |
+
"step": 700
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 5.6,
|
| 72 |
+
"eval_loss": 0.23278702795505524,
|
| 73 |
+
"eval_runtime": 211.6975,
|
| 74 |
+
"eval_samples_per_second": 18.106,
|
| 75 |
+
"eval_steps_per_second": 2.267,
|
| 76 |
+
"eval_wer": 0.1827395418584208,
|
| 77 |
+
"step": 750
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 5.97,
|
| 81 |
+
"learning_rate": 4.9973915963877904e-05,
|
| 82 |
+
"loss": 0.031,
|
| 83 |
+
"step": 800
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 6.72,
|
| 87 |
+
"learning_rate": 4.9970184342115084e-05,
|
| 88 |
+
"loss": 0.0287,
|
| 89 |
+
"step": 900
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 7.46,
|
| 93 |
+
"learning_rate": 4.996645272035227e-05,
|
| 94 |
+
"loss": 0.0281,
|
| 95 |
+
"step": 1000
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 7.46,
|
| 99 |
+
"eval_loss": 0.23425738513469696,
|
| 100 |
+
"eval_runtime": 209.1881,
|
| 101 |
+
"eval_samples_per_second": 18.323,
|
| 102 |
+
"eval_steps_per_second": 2.295,
|
| 103 |
+
"eval_wer": 0.18170536222141787,
|
| 104 |
+
"step": 1000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 8.21,
|
| 108 |
+
"learning_rate": 4.996272109858945e-05,
|
| 109 |
+
"loss": 0.0266,
|
| 110 |
+
"step": 1100
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 8.95,
|
| 114 |
+
"learning_rate": 4.995898947682663e-05,
|
| 115 |
+
"loss": 0.0263,
|
| 116 |
+
"step": 1200
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 9.33,
|
| 120 |
+
"eval_loss": 0.23851124942302704,
|
| 121 |
+
"eval_runtime": 208.1334,
|
| 122 |
+
"eval_samples_per_second": 18.416,
|
| 123 |
+
"eval_steps_per_second": 2.306,
|
| 124 |
+
"eval_wer": 0.17855111432855886,
|
| 125 |
+
"step": 1250
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 9.7,
|
| 129 |
+
"learning_rate": 4.995525785506381e-05,
|
| 130 |
+
"loss": 0.0265,
|
| 131 |
+
"step": 1300
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 10.45,
|
| 135 |
+
"learning_rate": 4.995152623330099e-05,
|
| 136 |
+
"loss": 0.0263,
|
| 137 |
+
"step": 1400
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 11.19,
|
| 141 |
+
"learning_rate": 4.994779461153818e-05,
|
| 142 |
+
"loss": 0.0262,
|
| 143 |
+
"step": 1500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 11.19,
|
| 147 |
+
"eval_loss": 0.2488125115633011,
|
| 148 |
+
"eval_runtime": 216.4749,
|
| 149 |
+
"eval_samples_per_second": 17.706,
|
| 150 |
+
"eval_steps_per_second": 2.217,
|
| 151 |
+
"eval_wer": 0.17699984487305445,
|
| 152 |
+
"step": 1500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 11.94,
|
| 156 |
+
"learning_rate": 4.994406298977536e-05,
|
| 157 |
+
"loss": 0.0237,
|
| 158 |
+
"step": 1600
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 12.69,
|
| 162 |
+
"learning_rate": 4.994033136801254e-05,
|
| 163 |
+
"loss": 0.0249,
|
| 164 |
+
"step": 1700
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 13.06,
|
| 168 |
+
"eval_loss": 0.261038601398468,
|
| 169 |
+
"eval_runtime": 219.4529,
|
| 170 |
+
"eval_samples_per_second": 17.466,
|
| 171 |
+
"eval_steps_per_second": 2.187,
|
| 172 |
+
"eval_wer": 0.17513832152644915,
|
| 173 |
+
"step": 1750
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 13.43,
|
| 177 |
+
"learning_rate": 4.9936599746249724e-05,
|
| 178 |
+
"loss": 0.0235,
|
| 179 |
+
"step": 1800
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 14.18,
|
| 183 |
+
"learning_rate": 4.9932868124486904e-05,
|
| 184 |
+
"loss": 0.0246,
|
| 185 |
+
"step": 1900
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 14.92,
|
| 189 |
+
"learning_rate": 4.992913650272409e-05,
|
| 190 |
+
"loss": 0.0227,
|
| 191 |
+
"step": 2000
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 14.92,
|
| 195 |
+
"eval_loss": 0.25547119975090027,
|
| 196 |
+
"eval_runtime": 215.7055,
|
| 197 |
+
"eval_samples_per_second": 17.77,
|
| 198 |
+
"eval_steps_per_second": 2.225,
|
| 199 |
+
"eval_wer": 0.17472464967164797,
|
| 200 |
+
"step": 2000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 15.67,
|
| 204 |
+
"learning_rate": 4.992540488096127e-05,
|
| 205 |
+
"loss": 0.0215,
|
| 206 |
+
"step": 2100
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 16.42,
|
| 210 |
+
"learning_rate": 4.992167325919845e-05,
|
| 211 |
+
"loss": 0.0218,
|
| 212 |
+
"step": 2200
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 16.79,
|
| 216 |
+
"eval_loss": 0.25752121210098267,
|
| 217 |
+
"eval_runtime": 207.5079,
|
| 218 |
+
"eval_samples_per_second": 18.472,
|
| 219 |
+
"eval_steps_per_second": 2.313,
|
| 220 |
+
"eval_wer": 0.17425926883499665,
|
| 221 |
+
"step": 2250
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 17.16,
|
| 225 |
+
"learning_rate": 4.991794163743563e-05,
|
| 226 |
+
"loss": 0.0213,
|
| 227 |
+
"step": 2300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 17.91,
|
| 231 |
+
"learning_rate": 4.991421001567281e-05,
|
| 232 |
+
"loss": 0.0212,
|
| 233 |
+
"step": 2400
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 18.66,
|
| 237 |
+
"learning_rate": 4.991047839390999e-05,
|
| 238 |
+
"loss": 0.0211,
|
| 239 |
+
"step": 2500
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 18.66,
|
| 243 |
+
"eval_loss": 0.2587244212627411,
|
| 244 |
+
"eval_runtime": 211.1091,
|
| 245 |
+
"eval_samples_per_second": 18.156,
|
| 246 |
+
"eval_steps_per_second": 2.274,
|
| 247 |
+
"eval_wer": 0.1761725011634521,
|
| 248 |
+
"step": 2500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 19.4,
|
| 252 |
+
"learning_rate": 4.990674677214718e-05,
|
| 253 |
+
"loss": 0.0201,
|
| 254 |
+
"step": 2600
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 20.15,
|
| 258 |
+
"learning_rate": 4.990301515038436e-05,
|
| 259 |
+
"loss": 0.0222,
|
| 260 |
+
"step": 2700
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 20.52,
|
| 264 |
+
"eval_loss": 0.26894259452819824,
|
| 265 |
+
"eval_runtime": 212.0938,
|
| 266 |
+
"eval_samples_per_second": 18.072,
|
| 267 |
+
"eval_steps_per_second": 2.263,
|
| 268 |
+
"eval_wer": 0.17451781374424738,
|
| 269 |
+
"step": 2750
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 20.89,
|
| 273 |
+
"learning_rate": 4.9899283528621545e-05,
|
| 274 |
+
"loss": 0.0207,
|
| 275 |
+
"step": 2800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 21.64,
|
| 279 |
+
"learning_rate": 4.9895551906858725e-05,
|
| 280 |
+
"loss": 0.0194,
|
| 281 |
+
"step": 2900
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 22.39,
|
| 285 |
+
"learning_rate": 4.9891820285095905e-05,
|
| 286 |
+
"loss": 0.021,
|
| 287 |
+
"step": 3000
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 22.39,
|
| 291 |
+
"eval_loss": 0.26362183690071106,
|
| 292 |
+
"eval_runtime": 204.8903,
|
| 293 |
+
"eval_samples_per_second": 18.708,
|
| 294 |
+
"eval_steps_per_second": 2.343,
|
| 295 |
+
"eval_wer": 0.1751900305082993,
|
| 296 |
+
"step": 3000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 23.13,
|
| 300 |
+
"learning_rate": 4.9888088663333085e-05,
|
| 301 |
+
"loss": 0.0216,
|
| 302 |
+
"step": 3100
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 23.88,
|
| 306 |
+
"learning_rate": 4.9884357041570265e-05,
|
| 307 |
+
"loss": 0.02,
|
| 308 |
+
"step": 3200
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 24.25,
|
| 312 |
+
"eval_loss": 0.2709738612174988,
|
| 313 |
+
"eval_runtime": 211.3483,
|
| 314 |
+
"eval_samples_per_second": 18.136,
|
| 315 |
+
"eval_steps_per_second": 2.271,
|
| 316 |
+
"eval_wer": 0.17513832152644915,
|
| 317 |
+
"step": 3250
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 24.63,
|
| 321 |
+
"learning_rate": 4.988062541980745e-05,
|
| 322 |
+
"loss": 0.02,
|
| 323 |
+
"step": 3300
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 25.37,
|
| 327 |
+
"learning_rate": 4.987689379804463e-05,
|
| 328 |
+
"loss": 0.0206,
|
| 329 |
+
"step": 3400
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 26.12,
|
| 333 |
+
"learning_rate": 4.987316217628181e-05,
|
| 334 |
+
"loss": 0.0195,
|
| 335 |
+
"step": 3500
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 26.12,
|
| 339 |
+
"eval_loss": 0.26540952920913696,
|
| 340 |
+
"eval_runtime": 209.7499,
|
| 341 |
+
"eval_samples_per_second": 18.274,
|
| 342 |
+
"eval_steps_per_second": 2.288,
|
| 343 |
+
"eval_wer": 0.17570712032680078,
|
| 344 |
+
"step": 3500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 26.86,
|
| 348 |
+
"learning_rate": 4.9869430554519e-05,
|
| 349 |
+
"loss": 0.0193,
|
| 350 |
+
"step": 3600
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 27.61,
|
| 354 |
+
"learning_rate": 4.986569893275618e-05,
|
| 355 |
+
"loss": 0.0197,
|
| 356 |
+
"step": 3700
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 27.98,
|
| 360 |
+
"eval_loss": 0.26861485838890076,
|
| 361 |
+
"eval_runtime": 224.5,
|
| 362 |
+
"eval_samples_per_second": 17.073,
|
| 363 |
+
"eval_steps_per_second": 2.138,
|
| 364 |
+
"eval_wer": 0.17482806763534825,
|
| 365 |
+
"step": 3750
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 28.36,
|
| 369 |
+
"learning_rate": 4.9861967310993365e-05,
|
| 370 |
+
"loss": 0.0198,
|
| 371 |
+
"step": 3800
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 29.1,
|
| 375 |
+
"learning_rate": 4.9858235689230545e-05,
|
| 376 |
+
"loss": 0.0199,
|
| 377 |
+
"step": 3900
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 29.85,
|
| 381 |
+
"learning_rate": 4.9854504067467725e-05,
|
| 382 |
+
"loss": 0.0187,
|
| 383 |
+
"step": 4000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 29.85,
|
| 387 |
+
"eval_loss": 0.2687038481235504,
|
| 388 |
+
"eval_runtime": 221.4533,
|
| 389 |
+
"eval_samples_per_second": 17.308,
|
| 390 |
+
"eval_steps_per_second": 2.168,
|
| 391 |
+
"eval_wer": 0.1751900305082993,
|
| 392 |
+
"step": 4000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 30.6,
|
| 396 |
+
"learning_rate": 4.9850772445704905e-05,
|
| 397 |
+
"loss": 0.019,
|
| 398 |
+
"step": 4100
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 31.34,
|
| 402 |
+
"learning_rate": 4.9847040823942085e-05,
|
| 403 |
+
"loss": 0.0188,
|
| 404 |
+
"step": 4200
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 31.72,
|
| 408 |
+
"eval_loss": 0.2625608444213867,
|
| 409 |
+
"eval_runtime": 222.8747,
|
| 410 |
+
"eval_samples_per_second": 17.198,
|
| 411 |
+
"eval_steps_per_second": 2.154,
|
| 412 |
+
"eval_wer": 0.1721392005791406,
|
| 413 |
+
"step": 4250
|
| 414 |
}
|
| 415 |
],
|
| 416 |
"max_steps": 1340000,
|
| 417 |
"num_train_epochs": 10000,
|
| 418 |
+
"total_flos": 9.643398954166528e+19,
|
| 419 |
"trial_name": null,
|
| 420 |
"trial_params": null
|
| 421 |
}
|