Model save
Browse files- last-checkpoint/config.json +0 -54
- last-checkpoint/optimizer.pt +0 -3
- last-checkpoint/pytorch_model.bin +0 -3
- last-checkpoint/rng_state.pth +0 -3
- last-checkpoint/scaler.pt +0 -3
- last-checkpoint/scheduler.pt +0 -3
- last-checkpoint/trainer_state.json +0 -544
- last-checkpoint/training_args.bin +0 -3
- runs/Oct24_08-39-52_00e58a194ed0/events.out.tfevents.1666601028.00e58a194ed0.23.0 +2 -2
last-checkpoint/config.json
DELETED
|
@@ -1,54 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"_name_or_path": "cedpsam/EleutherAI_gpt-neo-125M-stablediffionprompts",
|
| 3 |
-
"activation_function": "gelu_new",
|
| 4 |
-
"architectures": [
|
| 5 |
-
"GPTNeoForCausalLM"
|
| 6 |
-
],
|
| 7 |
-
"attention_dropout": 0,
|
| 8 |
-
"attention_layers": [
|
| 9 |
-
"global",
|
| 10 |
-
"local",
|
| 11 |
-
"global",
|
| 12 |
-
"local",
|
| 13 |
-
"global",
|
| 14 |
-
"local",
|
| 15 |
-
"global",
|
| 16 |
-
"local",
|
| 17 |
-
"global",
|
| 18 |
-
"local",
|
| 19 |
-
"global",
|
| 20 |
-
"local"
|
| 21 |
-
],
|
| 22 |
-
"attention_types": [
|
| 23 |
-
[
|
| 24 |
-
[
|
| 25 |
-
"global",
|
| 26 |
-
"local"
|
| 27 |
-
],
|
| 28 |
-
6
|
| 29 |
-
]
|
| 30 |
-
],
|
| 31 |
-
"bos_token_id": 50256,
|
| 32 |
-
"embed_dropout": 0,
|
| 33 |
-
"eos_token_id": 50256,
|
| 34 |
-
"gradient_checkpointing": false,
|
| 35 |
-
"hidden_size": 768,
|
| 36 |
-
"initializer_range": 0.02,
|
| 37 |
-
"intermediate_size": null,
|
| 38 |
-
"layer_norm_epsilon": 1e-05,
|
| 39 |
-
"max_position_embeddings": 2048,
|
| 40 |
-
"model_type": "gpt_neo",
|
| 41 |
-
"num_heads": 12,
|
| 42 |
-
"num_layers": 12,
|
| 43 |
-
"resid_dropout": 0,
|
| 44 |
-
"summary_activation": null,
|
| 45 |
-
"summary_first_dropout": 0.1,
|
| 46 |
-
"summary_proj_to_labels": true,
|
| 47 |
-
"summary_type": "cls_index",
|
| 48 |
-
"summary_use_proj": true,
|
| 49 |
-
"torch_dtype": "float32",
|
| 50 |
-
"transformers_version": "4.20.1",
|
| 51 |
-
"use_cache": true,
|
| 52 |
-
"vocab_size": 50257,
|
| 53 |
-
"window_size": 256
|
| 54 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
last-checkpoint/optimizer.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:58d7908b60759dc21ccb2c8e7e0912f4f93a76d680317cd5fec5ae118ab7f2c7
|
| 3 |
-
size 1001681601
|
|
|
|
|
|
|
|
|
last-checkpoint/pytorch_model.bin
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:cecc6aeb658ce5cddba59472bfbea3e76852b232a433f21fe1213ccd1ba34907
|
| 3 |
-
size 551185105
|
|
|
|
|
|
|
|
|
last-checkpoint/rng_state.pth
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:62851f63d26f31f0098aaf22bfc590da9d5dd492f2a149b0c502c13be7100a3d
|
| 3 |
-
size 14503
|
|
|
|
|
|
|
|
|
last-checkpoint/scaler.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:d9640f7e701d01460e354f09734e46a7a40d5d9a9364ed130984bf2e740abc77
|
| 3 |
-
size 559
|
|
|
|
|
|
|
|
|
last-checkpoint/scheduler.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:be5d33d56fa8110787b4a768e6ee5c083b024ac4d60c4bcbcaa263471474ae4f
|
| 3 |
-
size 623
|
|
|
|
|
|
|
|
|
last-checkpoint/trainer_state.json
DELETED
|
@@ -1,544 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_metric": null,
|
| 3 |
-
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.5968691500040696,
|
| 5 |
-
"global_step": 44000,
|
| 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": 9.886136363636363e-08,
|
| 13 |
-
"loss": 0.024,
|
| 14 |
-
"step": 500
|
| 15 |
-
},
|
| 16 |
-
{
|
| 17 |
-
"epoch": 0.01,
|
| 18 |
-
"learning_rate": 9.7725e-08,
|
| 19 |
-
"loss": 0.0269,
|
| 20 |
-
"step": 1000
|
| 21 |
-
},
|
| 22 |
-
{
|
| 23 |
-
"epoch": 0.02,
|
| 24 |
-
"learning_rate": 9.658863636363636e-08,
|
| 25 |
-
"loss": 0.0252,
|
| 26 |
-
"step": 1500
|
| 27 |
-
},
|
| 28 |
-
{
|
| 29 |
-
"epoch": 0.03,
|
| 30 |
-
"learning_rate": 9.545227272727273e-08,
|
| 31 |
-
"loss": 0.0258,
|
| 32 |
-
"step": 2000
|
| 33 |
-
},
|
| 34 |
-
{
|
| 35 |
-
"epoch": 0.03,
|
| 36 |
-
"learning_rate": 9.431590909090909e-08,
|
| 37 |
-
"loss": 0.0279,
|
| 38 |
-
"step": 2500
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.04,
|
| 42 |
-
"learning_rate": 9.317954545454546e-08,
|
| 43 |
-
"loss": 0.0234,
|
| 44 |
-
"step": 3000
|
| 45 |
-
},
|
| 46 |
-
{
|
| 47 |
-
"epoch": 0.05,
|
| 48 |
-
"learning_rate": 9.204318181818181e-08,
|
| 49 |
-
"loss": 0.0261,
|
| 50 |
-
"step": 3500
|
| 51 |
-
},
|
| 52 |
-
{
|
| 53 |
-
"epoch": 0.05,
|
| 54 |
-
"learning_rate": 9.090681818181817e-08,
|
| 55 |
-
"loss": 0.0256,
|
| 56 |
-
"step": 4000
|
| 57 |
-
},
|
| 58 |
-
{
|
| 59 |
-
"epoch": 0.06,
|
| 60 |
-
"learning_rate": 8.977045454545454e-08,
|
| 61 |
-
"loss": 0.0265,
|
| 62 |
-
"step": 4500
|
| 63 |
-
},
|
| 64 |
-
{
|
| 65 |
-
"epoch": 0.07,
|
| 66 |
-
"learning_rate": 8.86340909090909e-08,
|
| 67 |
-
"loss": 0.0244,
|
| 68 |
-
"step": 5000
|
| 69 |
-
},
|
| 70 |
-
{
|
| 71 |
-
"epoch": 0.07,
|
| 72 |
-
"learning_rate": 8.749772727272727e-08,
|
| 73 |
-
"loss": 0.0253,
|
| 74 |
-
"step": 5500
|
| 75 |
-
},
|
| 76 |
-
{
|
| 77 |
-
"epoch": 0.08,
|
| 78 |
-
"learning_rate": 8.636136363636363e-08,
|
| 79 |
-
"loss": 0.0258,
|
| 80 |
-
"step": 6000
|
| 81 |
-
},
|
| 82 |
-
{
|
| 83 |
-
"epoch": 0.09,
|
| 84 |
-
"learning_rate": 8.522727272727273e-08,
|
| 85 |
-
"loss": 0.0256,
|
| 86 |
-
"step": 6500
|
| 87 |
-
},
|
| 88 |
-
{
|
| 89 |
-
"epoch": 0.09,
|
| 90 |
-
"learning_rate": 8.409090909090909e-08,
|
| 91 |
-
"loss": 0.0251,
|
| 92 |
-
"step": 7000
|
| 93 |
-
},
|
| 94 |
-
{
|
| 95 |
-
"epoch": 0.1,
|
| 96 |
-
"learning_rate": 8.295454545454546e-08,
|
| 97 |
-
"loss": 0.026,
|
| 98 |
-
"step": 7500
|
| 99 |
-
},
|
| 100 |
-
{
|
| 101 |
-
"epoch": 0.11,
|
| 102 |
-
"learning_rate": 8.181818181818182e-08,
|
| 103 |
-
"loss": 0.0263,
|
| 104 |
-
"step": 8000
|
| 105 |
-
},
|
| 106 |
-
{
|
| 107 |
-
"epoch": 0.12,
|
| 108 |
-
"learning_rate": 8.068181818181817e-08,
|
| 109 |
-
"loss": 0.0261,
|
| 110 |
-
"step": 8500
|
| 111 |
-
},
|
| 112 |
-
{
|
| 113 |
-
"epoch": 0.12,
|
| 114 |
-
"learning_rate": 7.954772727272727e-08,
|
| 115 |
-
"loss": 0.0253,
|
| 116 |
-
"step": 9000
|
| 117 |
-
},
|
| 118 |
-
{
|
| 119 |
-
"epoch": 0.13,
|
| 120 |
-
"learning_rate": 7.841136363636363e-08,
|
| 121 |
-
"loss": 0.0256,
|
| 122 |
-
"step": 9500
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 0.14,
|
| 126 |
-
"learning_rate": 7.7275e-08,
|
| 127 |
-
"loss": 0.0254,
|
| 128 |
-
"step": 10000
|
| 129 |
-
},
|
| 130 |
-
{
|
| 131 |
-
"epoch": 0.14,
|
| 132 |
-
"learning_rate": 7.613863636363636e-08,
|
| 133 |
-
"loss": 0.0247,
|
| 134 |
-
"step": 10500
|
| 135 |
-
},
|
| 136 |
-
{
|
| 137 |
-
"epoch": 0.15,
|
| 138 |
-
"learning_rate": 7.500681818181818e-08,
|
| 139 |
-
"loss": 0.0266,
|
| 140 |
-
"step": 11000
|
| 141 |
-
},
|
| 142 |
-
{
|
| 143 |
-
"epoch": 0.16,
|
| 144 |
-
"learning_rate": 7.387045454545455e-08,
|
| 145 |
-
"loss": 0.026,
|
| 146 |
-
"step": 11500
|
| 147 |
-
},
|
| 148 |
-
{
|
| 149 |
-
"epoch": 0.16,
|
| 150 |
-
"learning_rate": 7.273409090909091e-08,
|
| 151 |
-
"loss": 0.0258,
|
| 152 |
-
"step": 12000
|
| 153 |
-
},
|
| 154 |
-
{
|
| 155 |
-
"epoch": 0.17,
|
| 156 |
-
"learning_rate": 7.159772727272728e-08,
|
| 157 |
-
"loss": 0.0255,
|
| 158 |
-
"step": 12500
|
| 159 |
-
},
|
| 160 |
-
{
|
| 161 |
-
"epoch": 0.18,
|
| 162 |
-
"learning_rate": 7.046136363636363e-08,
|
| 163 |
-
"loss": 0.0261,
|
| 164 |
-
"step": 13000
|
| 165 |
-
},
|
| 166 |
-
{
|
| 167 |
-
"epoch": 0.18,
|
| 168 |
-
"learning_rate": 6.9325e-08,
|
| 169 |
-
"loss": 0.0251,
|
| 170 |
-
"step": 13500
|
| 171 |
-
},
|
| 172 |
-
{
|
| 173 |
-
"epoch": 0.19,
|
| 174 |
-
"learning_rate": 6.818863636363636e-08,
|
| 175 |
-
"loss": 0.0247,
|
| 176 |
-
"step": 14000
|
| 177 |
-
},
|
| 178 |
-
{
|
| 179 |
-
"epoch": 0.2,
|
| 180 |
-
"learning_rate": 6.705454545454545e-08,
|
| 181 |
-
"loss": 0.0256,
|
| 182 |
-
"step": 14500
|
| 183 |
-
},
|
| 184 |
-
{
|
| 185 |
-
"epoch": 0.2,
|
| 186 |
-
"learning_rate": 6.591818181818182e-08,
|
| 187 |
-
"loss": 0.0253,
|
| 188 |
-
"step": 15000
|
| 189 |
-
},
|
| 190 |
-
{
|
| 191 |
-
"epoch": 0.21,
|
| 192 |
-
"learning_rate": 6.478181818181818e-08,
|
| 193 |
-
"loss": 0.0242,
|
| 194 |
-
"step": 15500
|
| 195 |
-
},
|
| 196 |
-
{
|
| 197 |
-
"epoch": 0.22,
|
| 198 |
-
"learning_rate": 6.364545454545455e-08,
|
| 199 |
-
"loss": 0.0246,
|
| 200 |
-
"step": 16000
|
| 201 |
-
},
|
| 202 |
-
{
|
| 203 |
-
"epoch": 0.22,
|
| 204 |
-
"learning_rate": 6.250909090909091e-08,
|
| 205 |
-
"loss": 0.0245,
|
| 206 |
-
"step": 16500
|
| 207 |
-
},
|
| 208 |
-
{
|
| 209 |
-
"epoch": 0.23,
|
| 210 |
-
"learning_rate": 6.137272727272726e-08,
|
| 211 |
-
"loss": 0.0249,
|
| 212 |
-
"step": 17000
|
| 213 |
-
},
|
| 214 |
-
{
|
| 215 |
-
"epoch": 0.24,
|
| 216 |
-
"learning_rate": 6.023636363636363e-08,
|
| 217 |
-
"loss": 0.0241,
|
| 218 |
-
"step": 17500
|
| 219 |
-
},
|
| 220 |
-
{
|
| 221 |
-
"epoch": 0.24,
|
| 222 |
-
"learning_rate": 5.909999999999999e-08,
|
| 223 |
-
"loss": 0.0246,
|
| 224 |
-
"step": 18000
|
| 225 |
-
},
|
| 226 |
-
{
|
| 227 |
-
"epoch": 0.25,
|
| 228 |
-
"learning_rate": 5.796363636363636e-08,
|
| 229 |
-
"loss": 0.0249,
|
| 230 |
-
"step": 18500
|
| 231 |
-
},
|
| 232 |
-
{
|
| 233 |
-
"epoch": 0.26,
|
| 234 |
-
"learning_rate": 5.6829545454545445e-08,
|
| 235 |
-
"loss": 0.0268,
|
| 236 |
-
"step": 19000
|
| 237 |
-
},
|
| 238 |
-
{
|
| 239 |
-
"epoch": 0.26,
|
| 240 |
-
"learning_rate": 5.569318181818181e-08,
|
| 241 |
-
"loss": 0.0255,
|
| 242 |
-
"step": 19500
|
| 243 |
-
},
|
| 244 |
-
{
|
| 245 |
-
"epoch": 0.27,
|
| 246 |
-
"learning_rate": 5.4556818181818175e-08,
|
| 247 |
-
"loss": 0.0255,
|
| 248 |
-
"step": 20000
|
| 249 |
-
},
|
| 250 |
-
{
|
| 251 |
-
"epoch": 0.28,
|
| 252 |
-
"learning_rate": 5.342045454545454e-08,
|
| 253 |
-
"loss": 0.027,
|
| 254 |
-
"step": 20500
|
| 255 |
-
},
|
| 256 |
-
{
|
| 257 |
-
"epoch": 0.28,
|
| 258 |
-
"learning_rate": 5.2284090909090905e-08,
|
| 259 |
-
"loss": 0.0252,
|
| 260 |
-
"step": 21000
|
| 261 |
-
},
|
| 262 |
-
{
|
| 263 |
-
"epoch": 0.29,
|
| 264 |
-
"learning_rate": 5.114772727272727e-08,
|
| 265 |
-
"loss": 0.0271,
|
| 266 |
-
"step": 21500
|
| 267 |
-
},
|
| 268 |
-
{
|
| 269 |
-
"epoch": 0.3,
|
| 270 |
-
"learning_rate": 5.0011363636363634e-08,
|
| 271 |
-
"loss": 0.0246,
|
| 272 |
-
"step": 22000
|
| 273 |
-
},
|
| 274 |
-
{
|
| 275 |
-
"epoch": 0.31,
|
| 276 |
-
"learning_rate": 4.8875e-08,
|
| 277 |
-
"loss": 0.0261,
|
| 278 |
-
"step": 22500
|
| 279 |
-
},
|
| 280 |
-
{
|
| 281 |
-
"epoch": 0.31,
|
| 282 |
-
"learning_rate": 4.7740909090909086e-08,
|
| 283 |
-
"loss": 0.0242,
|
| 284 |
-
"step": 23000
|
| 285 |
-
},
|
| 286 |
-
{
|
| 287 |
-
"epoch": 0.32,
|
| 288 |
-
"learning_rate": 4.660909090909091e-08,
|
| 289 |
-
"loss": 0.0248,
|
| 290 |
-
"step": 23500
|
| 291 |
-
},
|
| 292 |
-
{
|
| 293 |
-
"epoch": 0.33,
|
| 294 |
-
"learning_rate": 4.5472727272727267e-08,
|
| 295 |
-
"loss": 0.025,
|
| 296 |
-
"step": 24000
|
| 297 |
-
},
|
| 298 |
-
{
|
| 299 |
-
"epoch": 0.33,
|
| 300 |
-
"learning_rate": 4.433636363636363e-08,
|
| 301 |
-
"loss": 0.0252,
|
| 302 |
-
"step": 24500
|
| 303 |
-
},
|
| 304 |
-
{
|
| 305 |
-
"epoch": 0.34,
|
| 306 |
-
"learning_rate": 4.3199999999999996e-08,
|
| 307 |
-
"loss": 0.0251,
|
| 308 |
-
"step": 25000
|
| 309 |
-
},
|
| 310 |
-
{
|
| 311 |
-
"epoch": 0.35,
|
| 312 |
-
"learning_rate": 4.206363636363636e-08,
|
| 313 |
-
"loss": 0.0254,
|
| 314 |
-
"step": 25500
|
| 315 |
-
},
|
| 316 |
-
{
|
| 317 |
-
"epoch": 0.35,
|
| 318 |
-
"learning_rate": 4.092954545454545e-08,
|
| 319 |
-
"loss": 0.0261,
|
| 320 |
-
"step": 26000
|
| 321 |
-
},
|
| 322 |
-
{
|
| 323 |
-
"epoch": 0.36,
|
| 324 |
-
"learning_rate": 3.979318181818181e-08,
|
| 325 |
-
"loss": 0.0262,
|
| 326 |
-
"step": 26500
|
| 327 |
-
},
|
| 328 |
-
{
|
| 329 |
-
"epoch": 0.37,
|
| 330 |
-
"learning_rate": 3.865681818181818e-08,
|
| 331 |
-
"loss": 0.0254,
|
| 332 |
-
"step": 27000
|
| 333 |
-
},
|
| 334 |
-
{
|
| 335 |
-
"epoch": 0.37,
|
| 336 |
-
"learning_rate": 3.752045454545454e-08,
|
| 337 |
-
"loss": 0.0247,
|
| 338 |
-
"step": 27500
|
| 339 |
-
},
|
| 340 |
-
{
|
| 341 |
-
"epoch": 0.38,
|
| 342 |
-
"learning_rate": 3.6386363636363636e-08,
|
| 343 |
-
"loss": 0.0252,
|
| 344 |
-
"step": 28000
|
| 345 |
-
},
|
| 346 |
-
{
|
| 347 |
-
"epoch": 0.39,
|
| 348 |
-
"learning_rate": 3.5249999999999995e-08,
|
| 349 |
-
"loss": 0.025,
|
| 350 |
-
"step": 28500
|
| 351 |
-
},
|
| 352 |
-
{
|
| 353 |
-
"epoch": 0.39,
|
| 354 |
-
"learning_rate": 3.411363636363636e-08,
|
| 355 |
-
"loss": 0.0255,
|
| 356 |
-
"step": 29000
|
| 357 |
-
},
|
| 358 |
-
{
|
| 359 |
-
"epoch": 0.4,
|
| 360 |
-
"learning_rate": 3.2977272727272724e-08,
|
| 361 |
-
"loss": 0.0255,
|
| 362 |
-
"step": 29500
|
| 363 |
-
},
|
| 364 |
-
{
|
| 365 |
-
"epoch": 0.41,
|
| 366 |
-
"learning_rate": 3.184318181818182e-08,
|
| 367 |
-
"loss": 0.0271,
|
| 368 |
-
"step": 30000
|
| 369 |
-
},
|
| 370 |
-
{
|
| 371 |
-
"epoch": 0.41,
|
| 372 |
-
"learning_rate": 3.0706818181818176e-08,
|
| 373 |
-
"loss": 0.0259,
|
| 374 |
-
"step": 30500
|
| 375 |
-
},
|
| 376 |
-
{
|
| 377 |
-
"epoch": 0.42,
|
| 378 |
-
"learning_rate": 2.9570454545454544e-08,
|
| 379 |
-
"loss": 0.0245,
|
| 380 |
-
"step": 31000
|
| 381 |
-
},
|
| 382 |
-
{
|
| 383 |
-
"epoch": 0.43,
|
| 384 |
-
"learning_rate": 2.8434090909090906e-08,
|
| 385 |
-
"loss": 0.024,
|
| 386 |
-
"step": 31500
|
| 387 |
-
},
|
| 388 |
-
{
|
| 389 |
-
"epoch": 0.43,
|
| 390 |
-
"learning_rate": 2.729772727272727e-08,
|
| 391 |
-
"loss": 0.0257,
|
| 392 |
-
"step": 32000
|
| 393 |
-
},
|
| 394 |
-
{
|
| 395 |
-
"epoch": 0.44,
|
| 396 |
-
"learning_rate": 2.6161363636363636e-08,
|
| 397 |
-
"loss": 0.0256,
|
| 398 |
-
"step": 32500
|
| 399 |
-
},
|
| 400 |
-
{
|
| 401 |
-
"epoch": 0.45,
|
| 402 |
-
"learning_rate": 2.5024999999999997e-08,
|
| 403 |
-
"loss": 0.0256,
|
| 404 |
-
"step": 33000
|
| 405 |
-
},
|
| 406 |
-
{
|
| 407 |
-
"epoch": 0.45,
|
| 408 |
-
"learning_rate": 2.3888636363636362e-08,
|
| 409 |
-
"loss": 0.0245,
|
| 410 |
-
"step": 33500
|
| 411 |
-
},
|
| 412 |
-
{
|
| 413 |
-
"epoch": 0.46,
|
| 414 |
-
"learning_rate": 2.2754545454545453e-08,
|
| 415 |
-
"loss": 0.0271,
|
| 416 |
-
"step": 34000
|
| 417 |
-
},
|
| 418 |
-
{
|
| 419 |
-
"epoch": 0.47,
|
| 420 |
-
"learning_rate": 2.1618181818181817e-08,
|
| 421 |
-
"loss": 0.0278,
|
| 422 |
-
"step": 34500
|
| 423 |
-
},
|
| 424 |
-
{
|
| 425 |
-
"epoch": 0.47,
|
| 426 |
-
"learning_rate": 2.048181818181818e-08,
|
| 427 |
-
"loss": 0.0246,
|
| 428 |
-
"step": 35000
|
| 429 |
-
},
|
| 430 |
-
{
|
| 431 |
-
"epoch": 0.48,
|
| 432 |
-
"learning_rate": 1.9345454545454544e-08,
|
| 433 |
-
"loss": 0.0252,
|
| 434 |
-
"step": 35500
|
| 435 |
-
},
|
| 436 |
-
{
|
| 437 |
-
"epoch": 0.49,
|
| 438 |
-
"learning_rate": 1.820909090909091e-08,
|
| 439 |
-
"loss": 0.0265,
|
| 440 |
-
"step": 36000
|
| 441 |
-
},
|
| 442 |
-
{
|
| 443 |
-
"epoch": 0.5,
|
| 444 |
-
"learning_rate": 1.707272727272727e-08,
|
| 445 |
-
"loss": 0.0257,
|
| 446 |
-
"step": 36500
|
| 447 |
-
},
|
| 448 |
-
{
|
| 449 |
-
"epoch": 0.5,
|
| 450 |
-
"learning_rate": 1.5936363636363635e-08,
|
| 451 |
-
"loss": 0.026,
|
| 452 |
-
"step": 37000
|
| 453 |
-
},
|
| 454 |
-
{
|
| 455 |
-
"epoch": 0.51,
|
| 456 |
-
"learning_rate": 1.4799999999999999e-08,
|
| 457 |
-
"loss": 0.0258,
|
| 458 |
-
"step": 37500
|
| 459 |
-
},
|
| 460 |
-
{
|
| 461 |
-
"epoch": 0.52,
|
| 462 |
-
"learning_rate": 1.3663636363636363e-08,
|
| 463 |
-
"loss": 0.0258,
|
| 464 |
-
"step": 38000
|
| 465 |
-
},
|
| 466 |
-
{
|
| 467 |
-
"epoch": 0.52,
|
| 468 |
-
"learning_rate": 1.2527272727272728e-08,
|
| 469 |
-
"loss": 0.0261,
|
| 470 |
-
"step": 38500
|
| 471 |
-
},
|
| 472 |
-
{
|
| 473 |
-
"epoch": 0.53,
|
| 474 |
-
"learning_rate": 1.1393181818181817e-08,
|
| 475 |
-
"loss": 0.0259,
|
| 476 |
-
"step": 39000
|
| 477 |
-
},
|
| 478 |
-
{
|
| 479 |
-
"epoch": 0.54,
|
| 480 |
-
"learning_rate": 1.0259090909090909e-08,
|
| 481 |
-
"loss": 0.0258,
|
| 482 |
-
"step": 39500
|
| 483 |
-
},
|
| 484 |
-
{
|
| 485 |
-
"epoch": 0.54,
|
| 486 |
-
"learning_rate": 9.122727272727272e-09,
|
| 487 |
-
"loss": 0.0265,
|
| 488 |
-
"step": 40000
|
| 489 |
-
},
|
| 490 |
-
{
|
| 491 |
-
"epoch": 0.55,
|
| 492 |
-
"learning_rate": 7.986363636363635e-09,
|
| 493 |
-
"loss": 0.0255,
|
| 494 |
-
"step": 40500
|
| 495 |
-
},
|
| 496 |
-
{
|
| 497 |
-
"epoch": 0.56,
|
| 498 |
-
"learning_rate": 6.85e-09,
|
| 499 |
-
"loss": 0.0242,
|
| 500 |
-
"step": 41000
|
| 501 |
-
},
|
| 502 |
-
{
|
| 503 |
-
"epoch": 0.56,
|
| 504 |
-
"learning_rate": 5.71590909090909e-09,
|
| 505 |
-
"loss": 0.0256,
|
| 506 |
-
"step": 41500
|
| 507 |
-
},
|
| 508 |
-
{
|
| 509 |
-
"epoch": 0.57,
|
| 510 |
-
"learning_rate": 4.581818181818182e-09,
|
| 511 |
-
"loss": 0.0245,
|
| 512 |
-
"step": 42000
|
| 513 |
-
},
|
| 514 |
-
{
|
| 515 |
-
"epoch": 0.58,
|
| 516 |
-
"learning_rate": 3.445454545454545e-09,
|
| 517 |
-
"loss": 0.0263,
|
| 518 |
-
"step": 42500
|
| 519 |
-
},
|
| 520 |
-
{
|
| 521 |
-
"epoch": 0.58,
|
| 522 |
-
"learning_rate": 2.309090909090909e-09,
|
| 523 |
-
"loss": 0.0257,
|
| 524 |
-
"step": 43000
|
| 525 |
-
},
|
| 526 |
-
{
|
| 527 |
-
"epoch": 0.59,
|
| 528 |
-
"learning_rate": 1.1727272727272726e-09,
|
| 529 |
-
"loss": 0.0266,
|
| 530 |
-
"step": 43500
|
| 531 |
-
},
|
| 532 |
-
{
|
| 533 |
-
"epoch": 0.6,
|
| 534 |
-
"learning_rate": 3.636363636363636e-11,
|
| 535 |
-
"loss": 0.0235,
|
| 536 |
-
"step": 44000
|
| 537 |
-
}
|
| 538 |
-
],
|
| 539 |
-
"max_steps": 44000,
|
| 540 |
-
"num_train_epochs": 1,
|
| 541 |
-
"total_flos": 4.597453878932275e+16,
|
| 542 |
-
"trial_name": null,
|
| 543 |
-
"trial_params": null
|
| 544 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
last-checkpoint/training_args.bin
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:0a9be4165819bdb518e6ce2fb6609c1da8c8f0fe6577ac5717e2df558fc930ee
|
| 3 |
-
size 3439
|
|
|
|
|
|
|
|
|
runs/Oct24_08-39-52_00e58a194ed0/events.out.tfevents.1666601028.00e58a194ed0.23.0
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 sha256:4e876b0bec176409fc42285f0a423fa45792c5740d3abdd694f9cde9fd72adae
|
| 3 |
+
size 59990
|