Upload folder using huggingface_hub
Browse files
qwen3_06B.blocks.12.hook_resid_post.sae.sparsity40.mse0.001.kl0.01.physics10.exp8/history.json
CHANGED
|
@@ -1,427 +1,427 @@
|
|
| 1 |
{
|
| 2 |
"loss": [
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
4.
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
|
| 18 |
],
|
| 19 |
"recon_loss": [
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
0.
|
| 23 |
-
0.
|
| 24 |
-
0.
|
| 25 |
-
0.
|
| 26 |
-
0.
|
| 27 |
-
0.
|
| 28 |
-
0.
|
| 29 |
-
0.
|
| 30 |
-
0.
|
| 31 |
-
0.
|
| 32 |
-
0.
|
| 33 |
-
0.
|
| 34 |
-
0.
|
| 35 |
],
|
| 36 |
"l1_loss": [
|
| 37 |
-
0.
|
| 38 |
-
0.
|
| 39 |
-
0.
|
| 40 |
-
0.
|
| 41 |
-
0.
|
| 42 |
-
0.
|
| 43 |
-
0.
|
| 44 |
-
0.
|
| 45 |
-
0.
|
| 46 |
-
0.
|
| 47 |
-
0.
|
| 48 |
-
0.
|
| 49 |
-
0.
|
| 50 |
-
0.
|
| 51 |
-
0.
|
| 52 |
],
|
| 53 |
"sparsity": [
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
|
| 60 |
-
|
| 61 |
-
|
| 62 |
-
56.
|
| 63 |
-
|
| 64 |
-
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
|
| 68 |
-
|
| 69 |
],
|
| 70 |
"recon_contribution": [
|
| 71 |
-
|
| 72 |
-
|
| 73 |
-
0.
|
| 74 |
-
0.
|
| 75 |
-
0.
|
| 76 |
-
0.
|
| 77 |
-
0.
|
| 78 |
-
0.
|
| 79 |
-
0.
|
| 80 |
-
0.
|
| 81 |
-
0.
|
| 82 |
-
0.
|
| 83 |
-
0.
|
| 84 |
-
0.
|
| 85 |
-
0.
|
| 86 |
],
|
| 87 |
"l1_contribution": [
|
| 88 |
-
|
| 89 |
-
|
| 90 |
-
|
| 91 |
-
|
| 92 |
-
|
| 93 |
-
|
| 94 |
-
|
| 95 |
-
|
| 96 |
-
|
| 97 |
-
|
| 98 |
-
|
| 99 |
-
|
| 100 |
-
|
| 101 |
-
|
| 102 |
-
|
| 103 |
],
|
| 104 |
"total_post_layer_mse": [
|
| 105 |
-
|
| 106 |
-
|
| 107 |
-
|
| 108 |
-
|
| 109 |
-
|
| 110 |
-
|
| 111 |
-
|
| 112 |
-
|
| 113 |
-
|
| 114 |
-
|
| 115 |
-
|
| 116 |
-
|
| 117 |
-
|
| 118 |
-
|
| 119 |
-
|
| 120 |
],
|
| 121 |
"block_mse_contribution": [
|
| 122 |
-
3.
|
| 123 |
-
0.
|
| 124 |
-
0.
|
| 125 |
-
0.
|
| 126 |
-
0.
|
| 127 |
-
0.
|
| 128 |
-
0.
|
| 129 |
-
0.
|
| 130 |
-
0.
|
| 131 |
-
0.
|
| 132 |
-
0.
|
| 133 |
-
0.
|
| 134 |
-
0.
|
| 135 |
-
0.
|
| 136 |
-
0.
|
| 137 |
],
|
| 138 |
"13_mse": [
|
| 139 |
-
|
| 140 |
-
|
| 141 |
-
0.
|
| 142 |
-
0.
|
| 143 |
-
0.
|
| 144 |
-
0.
|
| 145 |
-
0.
|
| 146 |
-
0.
|
| 147 |
-
0.
|
| 148 |
-
0.
|
| 149 |
-
0.
|
| 150 |
-
0.
|
| 151 |
-
0.
|
| 152 |
-
0.
|
| 153 |
-
0.
|
| 154 |
],
|
| 155 |
"14_mse": [
|
| 156 |
-
|
| 157 |
-
1.
|
| 158 |
-
0.
|
| 159 |
-
0.
|
| 160 |
-
0.
|
| 161 |
-
0.
|
| 162 |
-
0.
|
| 163 |
-
0.
|
| 164 |
-
0.
|
| 165 |
-
0.
|
| 166 |
-
0.
|
| 167 |
-
0.
|
| 168 |
-
0.
|
| 169 |
-
0.
|
| 170 |
-
0.
|
| 171 |
],
|
| 172 |
"15_mse": [
|
| 173 |
-
|
| 174 |
-
1.
|
| 175 |
-
0.
|
| 176 |
-
0.
|
| 177 |
-
0.
|
| 178 |
-
0.
|
| 179 |
-
0.
|
| 180 |
-
0.
|
| 181 |
-
0.
|
| 182 |
-
0.
|
| 183 |
-
0.
|
| 184 |
-
0.
|
| 185 |
-
0.
|
| 186 |
-
0.
|
| 187 |
-
0.
|
| 188 |
],
|
| 189 |
"16_mse": [
|
| 190 |
-
|
| 191 |
-
|
| 192 |
-
1.
|
| 193 |
-
1.
|
| 194 |
-
0.
|
| 195 |
-
0.
|
| 196 |
-
0.
|
| 197 |
-
0.
|
| 198 |
-
0.
|
| 199 |
-
0.
|
| 200 |
-
0.
|
| 201 |
-
0.
|
| 202 |
-
0.
|
| 203 |
-
0.
|
| 204 |
-
0.
|
| 205 |
],
|
| 206 |
"17_mse": [
|
| 207 |
-
|
| 208 |
-
2.
|
| 209 |
-
1.
|
| 210 |
-
1.
|
| 211 |
-
1.
|
| 212 |
-
1.
|
| 213 |
-
1.
|
| 214 |
-
|
| 215 |
-
|
| 216 |
-
0.
|
| 217 |
-
0.
|
| 218 |
-
0.
|
| 219 |
-
0.
|
| 220 |
-
0.
|
| 221 |
-
0.
|
| 222 |
],
|
| 223 |
"18_mse": [
|
| 224 |
-
|
| 225 |
-
3.
|
| 226 |
-
2.
|
| 227 |
-
1.
|
| 228 |
-
1.
|
| 229 |
-
1.
|
| 230 |
-
1.
|
| 231 |
-
1.
|
| 232 |
-
1.
|
| 233 |
-
1.
|
| 234 |
-
1.
|
| 235 |
-
1.
|
| 236 |
-
1.
|
| 237 |
-
1.
|
| 238 |
-
1.
|
| 239 |
],
|
| 240 |
"19_mse": [
|
| 241 |
-
|
| 242 |
-
4.
|
| 243 |
-
|
| 244 |
-
2.
|
| 245 |
-
|
| 246 |
-
|
| 247 |
-
|
| 248 |
-
1.
|
| 249 |
-
1.
|
| 250 |
-
1.
|
| 251 |
-
1.
|
| 252 |
-
1.
|
| 253 |
-
1.
|
| 254 |
-
1.
|
| 255 |
-
1.
|
| 256 |
],
|
| 257 |
"20_mse": [
|
| 258 |
-
|
| 259 |
-
5.
|
| 260 |
-
|
| 261 |
-
3.
|
| 262 |
-
|
| 263 |
-
|
| 264 |
-
2.
|
| 265 |
-
2.
|
| 266 |
-
2.
|
| 267 |
-
2.
|
| 268 |
-
|
| 269 |
-
2.
|
| 270 |
-
|
| 271 |
-
2.
|
| 272 |
-
2.
|
| 273 |
],
|
| 274 |
"21_mse": [
|
| 275 |
-
|
| 276 |
-
7.
|
| 277 |
-
|
| 278 |
-
|
| 279 |
-
|
| 280 |
-
|
| 281 |
-
3.
|
| 282 |
-
3.
|
| 283 |
-
3.
|
| 284 |
-
|
| 285 |
-
|
| 286 |
-
|
| 287 |
-
|
| 288 |
-
|
| 289 |
-
|
| 290 |
],
|
| 291 |
"22_mse": [
|
| 292 |
-
|
| 293 |
-
8.
|
| 294 |
-
|
| 295 |
-
|
| 296 |
-
|
| 297 |
-
|
| 298 |
-
4.
|
| 299 |
-
|
| 300 |
-
|
| 301 |
-
|
| 302 |
-
|
| 303 |
-
|
| 304 |
-
|
| 305 |
-
|
| 306 |
-
3.
|
| 307 |
],
|
| 308 |
"23_mse": [
|
| 309 |
-
|
| 310 |
-
11.
|
| 311 |
-
|
| 312 |
-
|
| 313 |
-
|
| 314 |
-
|
| 315 |
-
|
| 316 |
-
|
| 317 |
-
|
| 318 |
-
|
| 319 |
-
|
| 320 |
-
|
| 321 |
-
|
| 322 |
-
|
| 323 |
-
4.
|
| 324 |
],
|
| 325 |
"24_mse": [
|
| 326 |
-
|
| 327 |
-
14.
|
| 328 |
-
|
| 329 |
-
|
| 330 |
-
|
| 331 |
-
|
| 332 |
-
|
| 333 |
-
|
| 334 |
-
|
| 335 |
-
|
| 336 |
-
|
| 337 |
-
|
| 338 |
-
|
| 339 |
-
|
| 340 |
-
|
| 341 |
],
|
| 342 |
"25_mse": [
|
| 343 |
-
|
| 344 |
-
|
| 345 |
-
|
| 346 |
-
|
| 347 |
-
|
| 348 |
-
|
| 349 |
-
|
| 350 |
-
|
| 351 |
-
|
| 352 |
-
|
| 353 |
-
|
| 354 |
-
|
| 355 |
-
|
| 356 |
-
|
| 357 |
-
|
| 358 |
],
|
| 359 |
"26_mse": [
|
| 360 |
-
|
| 361 |
-
|
| 362 |
-
|
| 363 |
-
|
| 364 |
-
|
| 365 |
-
|
| 366 |
-
|
| 367 |
-
|
| 368 |
-
|
| 369 |
-
|
| 370 |
-
|
| 371 |
-
|
| 372 |
-
|
| 373 |
-
8.
|
| 374 |
-
|
| 375 |
],
|
| 376 |
"27_mse": [
|
| 377 |
-
|
| 378 |
-
|
| 379 |
-
|
| 380 |
-
|
| 381 |
-
|
| 382 |
-
|
| 383 |
-
|
| 384 |
-
|
| 385 |
-
|
| 386 |
-
|
| 387 |
-
|
| 388 |
-
|
| 389 |
-
|
| 390 |
-
|
| 391 |
-
|
| 392 |
],
|
| 393 |
"logit_kl": [
|
| 394 |
-
|
| 395 |
-
|
| 396 |
-
|
| 397 |
-
|
| 398 |
-
|
| 399 |
-
|
| 400 |
-
|
| 401 |
-
|
| 402 |
-
|
| 403 |
-
|
| 404 |
-
|
| 405 |
-
|
| 406 |
-
|
| 407 |
-
|
| 408 |
-
|
| 409 |
],
|
| 410 |
"kl_contribution": [
|
| 411 |
-
|
| 412 |
-
0.
|
| 413 |
-
0.
|
| 414 |
-
0.
|
| 415 |
-
0.
|
| 416 |
-
0.
|
| 417 |
-
0.
|
| 418 |
-
0.
|
| 419 |
-
0.
|
| 420 |
-
0.
|
| 421 |
-
0.
|
| 422 |
-
0.
|
| 423 |
-
0.
|
| 424 |
-
0.
|
| 425 |
-
0.
|
| 426 |
]
|
| 427 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"loss": [
|
| 3 |
+
132.40072370830336,
|
| 4 |
+
15.898285371916634,
|
| 5 |
+
4.808860845135567,
|
| 6 |
+
4.2647794995989114,
|
| 7 |
+
3.7614752946043373,
|
| 8 |
+
3.5804317333644495,
|
| 9 |
+
3.621848736938677,
|
| 10 |
+
3.311293805452218,
|
| 11 |
+
3.3564420907120955,
|
| 12 |
+
3.2430825179680847,
|
| 13 |
+
3.1446165334909484,
|
| 14 |
+
3.166066134334507,
|
| 15 |
+
3.145363519962569,
|
| 16 |
+
3.1751417849296915,
|
| 17 |
+
3.1040466840105845
|
| 18 |
],
|
| 19 |
"recon_loss": [
|
| 20 |
+
91.53491892021401,
|
| 21 |
+
1.127020146837808,
|
| 22 |
+
0.5399636414490248,
|
| 23 |
+
0.4608159236665955,
|
| 24 |
+
0.3498118930965438,
|
| 25 |
+
0.3483069581644876,
|
| 26 |
+
0.38624757724372966,
|
| 27 |
+
0.3225517376585114,
|
| 28 |
+
0.3369383092661549,
|
| 29 |
+
0.30339636519215163,
|
| 30 |
+
0.29119090242941575,
|
| 31 |
+
0.31196532981064085,
|
| 32 |
+
0.307742903387188,
|
| 33 |
+
0.32156725171813394,
|
| 34 |
+
0.30149288374678535
|
| 35 |
],
|
| 36 |
"l1_loss": [
|
| 37 |
+
0.8785219717966882,
|
| 38 |
+
0.3300659595603557,
|
| 39 |
+
0.08405589065829613,
|
| 40 |
+
0.07463387472737104,
|
| 41 |
+
0.0716574577810733,
|
| 42 |
+
0.06664694049500657,
|
| 43 |
+
0.06470343093142698,
|
| 44 |
+
0.06195041967934012,
|
| 45 |
+
0.061602180788671614,
|
| 46 |
+
0.0611304106064757,
|
| 47 |
+
0.0603844676152768,
|
| 48 |
+
0.05863903490896512,
|
| 49 |
+
0.05944218401818123,
|
| 50 |
+
0.05809492428079808,
|
| 51 |
+
0.05800197750357981
|
| 52 |
],
|
| 53 |
"sparsity": [
|
| 54 |
+
1057.4835558439556,
|
| 55 |
+
407.9275162202075,
|
| 56 |
+
162.19449604005743,
|
| 57 |
+
119.2390982262174,
|
| 58 |
+
95.5933568854081,
|
| 59 |
+
82.25088138150093,
|
| 60 |
+
70.82447925366853,
|
| 61 |
+
62.93163240762581,
|
| 62 |
+
56.17579608573053,
|
| 63 |
+
50.73961696051117,
|
| 64 |
+
46.83539435020963,
|
| 65 |
+
43.90063462938581,
|
| 66 |
+
40.67231744034846,
|
| 67 |
+
39.208685760211225,
|
| 68 |
+
37.08526147397837
|
| 69 |
],
|
| 70 |
"recon_contribution": [
|
| 71 |
+
91.53491892021401,
|
| 72 |
+
1.127020146837808,
|
| 73 |
+
0.5399636414490248,
|
| 74 |
+
0.4608159236665955,
|
| 75 |
+
0.3498118930965438,
|
| 76 |
+
0.3483069581644876,
|
| 77 |
+
0.38624757724372966,
|
| 78 |
+
0.3225517376585114,
|
| 79 |
+
0.3369383092661549,
|
| 80 |
+
0.30339636519215163,
|
| 81 |
+
0.29119090242941575,
|
| 82 |
+
0.31196532981064085,
|
| 83 |
+
0.307742903387188,
|
| 84 |
+
0.32156725171813394,
|
| 85 |
+
0.30149288374678535
|
| 86 |
],
|
| 87 |
"l1_contribution": [
|
| 88 |
+
35.14087875982872,
|
| 89 |
+
13.202638413673057,
|
| 90 |
+
3.3622356299170875,
|
| 91 |
+
2.9853549962653254,
|
| 92 |
+
2.8662983100665245,
|
| 93 |
+
2.6658776234415242,
|
| 94 |
+
2.5881372282379553,
|
| 95 |
+
2.478016788350012,
|
| 96 |
+
2.4640872312667677,
|
| 97 |
+
2.445216426723882,
|
| 98 |
+
2.4153787063476733,
|
| 99 |
+
2.345561395238217,
|
| 100 |
+
2.3776873603351136,
|
| 101 |
+
2.3237969747611453,
|
| 102 |
+
2.3200791032242596
|
| 103 |
],
|
| 104 |
"total_post_layer_mse": [
|
| 105 |
+
3768.518025380776,
|
| 106 |
+
634.3760369014472,
|
| 107 |
+
281.5267423777876,
|
| 108 |
+
257.9137649592152,
|
| 109 |
+
121.86403825979932,
|
| 110 |
+
132.82595429743142,
|
| 111 |
+
182.3500442832036,
|
| 112 |
+
120.80600859516098,
|
| 113 |
+
148.72142278666334,
|
| 114 |
+
119.53372874082928,
|
| 115 |
+
96.44286261605365,
|
| 116 |
+
132.2072357783528,
|
| 117 |
+
119.37015544875224,
|
| 118 |
+
140.67416488754571,
|
| 119 |
+
125.15249402512957
|
| 120 |
],
|
| 121 |
"block_mse_contribution": [
|
| 122 |
+
3.768518193868248,
|
| 123 |
+
0.6343760687253469,
|
| 124 |
+
0.28152675526179427,
|
| 125 |
+
0.257913777174597,
|
| 126 |
+
0.12186404388013675,
|
| 127 |
+
0.13282596073111194,
|
| 128 |
+
0.18235005254360945,
|
| 129 |
+
0.12080601431375555,
|
| 130 |
+
0.14872143043109295,
|
| 131 |
+
0.1195337344229062,
|
| 132 |
+
0.09644286709981384,
|
| 133 |
+
0.1322072420359626,
|
| 134 |
+
0.11937016143081264,
|
| 135 |
+
0.1406741713079483,
|
| 136 |
+
0.12515250030259253
|
| 137 |
],
|
| 138 |
"13_mse": [
|
| 139 |
+
92.58617470228583,
|
| 140 |
+
1.424407343443175,
|
| 141 |
+
0.6843610000341458,
|
| 142 |
+
0.5639535150581733,
|
| 143 |
+
0.42544728976891455,
|
| 144 |
+
0.4171061430658613,
|
| 145 |
+
0.45088799185770795,
|
| 146 |
+
0.3778063368080254,
|
| 147 |
+
0.3880942987096041,
|
| 148 |
+
0.3534062831921685,
|
| 149 |
+
0.3343137366543139,
|
| 150 |
+
0.35508942990598824,
|
| 151 |
+
0.347726347863226,
|
| 152 |
+
0.36387658421706437,
|
| 153 |
+
0.34385311766002413
|
| 154 |
],
|
| 155 |
"14_mse": [
|
| 156 |
+
92.43770923874432,
|
| 157 |
+
1.4219669226865124,
|
| 158 |
+
0.7184918425361017,
|
| 159 |
+
0.603204586003956,
|
| 160 |
+
0.4554870707872219,
|
| 161 |
+
0.4487679816950533,
|
| 162 |
+
0.48403634332624595,
|
| 163 |
+
0.40736286913541925,
|
| 164 |
+
0.41743570364507515,
|
| 165 |
+
0.38230037711616743,
|
| 166 |
+
0.3595608007863052,
|
| 167 |
+
0.3827579479692574,
|
| 168 |
+
0.3731682823788851,
|
| 169 |
+
0.3925049804655233,
|
| 170 |
+
0.3724479387353237
|
| 171 |
],
|
| 172 |
"15_mse": [
|
| 173 |
+
92.73112112597416,
|
| 174 |
+
1.5456405208074957,
|
| 175 |
+
0.8291421245811577,
|
| 176 |
+
0.696399899577736,
|
| 177 |
+
0.530499174966848,
|
| 178 |
+
0.5212662321956534,
|
| 179 |
+
0.5569074418311729,
|
| 180 |
+
0.47336215733137343,
|
| 181 |
+
0.48270386916802344,
|
| 182 |
+
0.44751030139457015,
|
| 183 |
+
0.41891130648161234,
|
| 184 |
+
0.44354036226308435,
|
| 185 |
+
0.43207019791567236,
|
| 186 |
+
0.4540229000543293,
|
| 187 |
+
0.4348179030239134
|
| 188 |
],
|
| 189 |
"16_mse": [
|
| 190 |
+
93.72751782800918,
|
| 191 |
+
2.088347224812759,
|
| 192 |
+
1.2250121937210399,
|
| 193 |
+
1.0437956295515363,
|
| 194 |
+
0.8037037549162269,
|
| 195 |
+
0.7893119821871134,
|
| 196 |
+
0.831803834527955,
|
| 197 |
+
0.714540472828356,
|
| 198 |
+
0.716447467754658,
|
| 199 |
+
0.6794583469181132,
|
| 200 |
+
0.628354791299741,
|
| 201 |
+
0.6565977841391599,
|
| 202 |
+
0.6375507835606883,
|
| 203 |
+
0.6720199679073534,
|
| 204 |
+
0.6464301565087828
|
| 205 |
],
|
| 206 |
"17_mse": [
|
| 207 |
+
94.47222283818668,
|
| 208 |
+
2.4767478825454425,
|
| 209 |
+
1.5378725349454951,
|
| 210 |
+
1.3074736180610227,
|
| 211 |
+
1.0276811555364078,
|
| 212 |
+
1.009038458193155,
|
| 213 |
+
1.052105562579363,
|
| 214 |
+
0.9169551736878273,
|
| 215 |
+
0.9124030157139427,
|
| 216 |
+
0.8749958547882568,
|
| 217 |
+
0.8088635177092445,
|
| 218 |
+
0.8403520140432774,
|
| 219 |
+
0.8132623561791011,
|
| 220 |
+
0.8571708014136866,
|
| 221 |
+
0.8309605762474519
|
| 222 |
],
|
| 223 |
"18_mse": [
|
| 224 |
+
96.62282840739516,
|
| 225 |
+
3.313168379597198,
|
| 226 |
+
2.1050605021025004,
|
| 227 |
+
1.7885640088776897,
|
| 228 |
+
1.4142655618208693,
|
| 229 |
+
1.3870145925005575,
|
| 230 |
+
1.4408194030586041,
|
| 231 |
+
1.2563351908124478,
|
| 232 |
+
1.2509042515342397,
|
| 233 |
+
1.2091203721842372,
|
| 234 |
+
1.1067663614911245,
|
| 235 |
+
1.1532457063072605,
|
| 236 |
+
1.1120133261035259,
|
| 237 |
+
1.1673076464269394,
|
| 238 |
+
1.144940797995804
|
| 239 |
],
|
| 240 |
"19_mse": [
|
| 241 |
+
98.12952686610974,
|
| 242 |
+
4.21422166842267,
|
| 243 |
+
2.7761693278649697,
|
| 244 |
+
2.3772155705251192,
|
| 245 |
+
1.9103159030577295,
|
| 246 |
+
1.8712453797347564,
|
| 247 |
+
1.930497253747811,
|
| 248 |
+
1.7012208476102442,
|
| 249 |
+
1.6912268557046588,
|
| 250 |
+
1.6499157534506088,
|
| 251 |
+
1.5067807480804902,
|
| 252 |
+
1.5715877149338113,
|
| 253 |
+
1.50961379762879,
|
| 254 |
+
1.5925079682715853,
|
| 255 |
+
1.569648396924026
|
| 256 |
],
|
| 257 |
"20_mse": [
|
| 258 |
+
100.86349562774028,
|
| 259 |
+
5.2358217391752655,
|
| 260 |
+
3.5889427751526797,
|
| 261 |
+
3.084748203144934,
|
| 262 |
+
2.5231203943266904,
|
| 263 |
+
2.466466988836016,
|
| 264 |
+
2.515812956300893,
|
| 265 |
+
2.2489302400359534,
|
| 266 |
+
2.2176499926954283,
|
| 267 |
+
2.1649505195761085,
|
| 268 |
+
1.996861229713698,
|
| 269 |
+
2.0641328773104157,
|
| 270 |
+
1.9832459145918824,
|
| 271 |
+
2.0868534244092785,
|
| 272 |
+
2.056514573276491
|
| 273 |
],
|
| 274 |
"21_mse": [
|
| 275 |
+
106.23677720521626,
|
| 276 |
+
7.429197506797045,
|
| 277 |
+
4.934605420083928,
|
| 278 |
+
4.241325343461861,
|
| 279 |
+
3.4353638143467724,
|
| 280 |
+
3.3451044389179776,
|
| 281 |
+
3.3986762430434836,
|
| 282 |
+
3.03673449674047,
|
| 283 |
+
3.0061623476501693,
|
| 284 |
+
2.8840246505307077,
|
| 285 |
+
2.644449479597852,
|
| 286 |
+
2.7081336916837477,
|
| 287 |
+
2.554877420117084,
|
| 288 |
+
2.698225707936108,
|
| 289 |
+
2.655916364121258
|
| 290 |
],
|
| 291 |
"22_mse": [
|
| 292 |
+
111.03444346807953,
|
| 293 |
+
8.813722875781526,
|
| 294 |
+
5.975814478737967,
|
| 295 |
+
5.13371875859741,
|
| 296 |
+
4.210649632869806,
|
| 297 |
+
4.100004742916365,
|
| 298 |
+
4.144820572738361,
|
| 299 |
+
3.7331808097380446,
|
| 300 |
+
3.6716954000014113,
|
| 301 |
+
3.5411856093801055,
|
| 302 |
+
3.2662032493075035,
|
| 303 |
+
3.32891203138165,
|
| 304 |
+
3.150659145269179,
|
| 305 |
+
3.3271091728282154,
|
| 306 |
+
3.274116275005771
|
| 307 |
],
|
| 308 |
"23_mse": [
|
| 309 |
+
118.86912513317021,
|
| 310 |
+
11.185627614645133,
|
| 311 |
+
7.6775109463168265,
|
| 312 |
+
6.536852415342977,
|
| 313 |
+
5.326670237053606,
|
| 314 |
+
5.195521380668296,
|
| 315 |
+
5.26501534845596,
|
| 316 |
+
4.727612521415367,
|
| 317 |
+
4.681196536336627,
|
| 318 |
+
4.525898637628197,
|
| 319 |
+
4.15481556268563,
|
| 320 |
+
4.256077368456618,
|
| 321 |
+
4.0400548922388175,
|
| 322 |
+
4.312153929158261,
|
| 323 |
+
4.251093875196643
|
| 324 |
],
|
| 325 |
"24_mse": [
|
| 326 |
+
130.50960404173773,
|
| 327 |
+
14.531528067768068,
|
| 328 |
+
9.785240420721527,
|
| 329 |
+
8.324993122789197,
|
| 330 |
+
6.708341277631602,
|
| 331 |
+
6.538236614456751,
|
| 332 |
+
6.6201859363039635,
|
| 333 |
+
5.944424200775032,
|
| 334 |
+
5.905932158455813,
|
| 335 |
+
5.711922184865277,
|
| 336 |
+
5.222949343516414,
|
| 337 |
+
5.336188742092678,
|
| 338 |
+
5.064628631548774,
|
| 339 |
+
5.40855013876033,
|
| 340 |
+
5.332810629579358
|
| 341 |
],
|
| 342 |
"25_mse": [
|
| 343 |
+
149.70575552775446,
|
| 344 |
+
18.781842669149988,
|
| 345 |
+
12.454813014295764,
|
| 346 |
+
10.676898454364977,
|
| 347 |
+
8.50971511790627,
|
| 348 |
+
8.30757270719772,
|
| 349 |
+
8.442735458675184,
|
| 350 |
+
7.540686890594941,
|
| 351 |
+
7.51053103647734,
|
| 352 |
+
7.272217981797412,
|
| 353 |
+
6.6091292926243375,
|
| 354 |
+
6.806205828387038,
|
| 355 |
+
6.429451408242821,
|
| 356 |
+
6.961337288519494,
|
| 357 |
+
6.801972677833156
|
| 358 |
],
|
| 359 |
"26_mse": [
|
| 360 |
+
162.7475875983561,
|
| 361 |
+
21.965660794336994,
|
| 362 |
+
14.38764136536677,
|
| 363 |
+
12.350393969313542,
|
| 364 |
+
9.63491886361201,
|
| 365 |
+
9.44224236782332,
|
| 366 |
+
9.776316755696348,
|
| 367 |
+
8.552372195666894,
|
| 368 |
+
8.593469288116111,
|
| 369 |
+
8.364798094096937,
|
| 370 |
+
7.4954372115601275,
|
| 371 |
+
7.8521559579031805,
|
| 372 |
+
7.360022383524959,
|
| 373 |
+
8.127834703689231,
|
| 374 |
+
7.897967291953869
|
| 375 |
],
|
| 376 |
"27_mse": [
|
| 377 |
+
2227.8441357720167,
|
| 378 |
+
529.9481356914779,
|
| 379 |
+
212.84606443132674,
|
| 380 |
+
199.18422786454508,
|
| 381 |
+
74.94785901119835,
|
| 382 |
+
86.98705428704284,
|
| 383 |
+
135.43942318106056,
|
| 384 |
+
79.17448419198058,
|
| 385 |
+
107.27557056470025,
|
| 386 |
+
79.47202377391041,
|
| 387 |
+
59.889465984545254,
|
| 388 |
+
94.45225832157566,
|
| 389 |
+
83.56181056158883,
|
| 390 |
+
102.2526896734883,
|
| 391 |
+
87.5390034510677
|
| 392 |
],
|
| 393 |
"logit_kl": [
|
| 394 |
+
195.64052740971846,
|
| 395 |
+
93.42507447694477,
|
| 396 |
+
62.51348275349552,
|
| 397 |
+
56.06948348454067,
|
| 398 |
+
42.35010630026796,
|
| 399 |
+
43.34211874008179,
|
| 400 |
+
46.51138848828194,
|
| 401 |
+
38.99192340750443,
|
| 402 |
+
40.66951527989897,
|
| 403 |
+
37.49360173806212,
|
| 404 |
+
34.16040521277521,
|
| 405 |
+
37.63321834937074,
|
| 406 |
+
34.05630800777808,
|
| 407 |
+
38.910339645873336,
|
| 408 |
+
35.732220771617456
|
| 409 |
],
|
| 410 |
"kl_contribution": [
|
| 411 |
+
1.9564052279058255,
|
| 412 |
+
0.9342507238226726,
|
| 413 |
+
0.6251348145586207,
|
| 414 |
+
0.5606948222993011,
|
| 415 |
+
0.42350105397907417,
|
| 416 |
+
0.433421177971632,
|
| 417 |
+
0.4651138730403176,
|
| 418 |
+
0.3899192253505825,
|
| 419 |
+
0.40669514345271246,
|
| 420 |
+
0.37493600797160226,
|
| 421 |
+
0.34160404475895984,
|
| 422 |
+
0.3763321751266494,
|
| 423 |
+
0.34056307274596137,
|
| 424 |
+
0.3891033879236171,
|
| 425 |
+
0.3573221996762699
|
| 426 |
]
|
| 427 |
}
|
qwen3_06B.blocks.12.hook_resid_post.sae.sparsity40.mse0.001.kl0.01.physics10.exp8/weights.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 67148125
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:687d03590ec005360ddc659688f1136b454240d73d02ca5312ee3e101e5d93ec
|
| 3 |
size 67148125
|