Delete folder add_sub_baseline_10K with huggingface_hub
Browse files
add_sub_baseline_10K/config.json
DELETED
|
@@ -1,37 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"SorlModelWrapper"
|
| 4 |
-
],
|
| 5 |
-
"attention_bias": false,
|
| 6 |
-
"attention_dropout": 0.0,
|
| 7 |
-
"bos_token_id": null,
|
| 8 |
-
"dtype": "float32",
|
| 9 |
-
"eos_token_id": null,
|
| 10 |
-
"head_dim": 128,
|
| 11 |
-
"hidden_act": "silu",
|
| 12 |
-
"hidden_size": 510,
|
| 13 |
-
"initializer_range": 0.02,
|
| 14 |
-
"intermediate_size": 2040,
|
| 15 |
-
"layer_types": [
|
| 16 |
-
"full_attention",
|
| 17 |
-
"full_attention"
|
| 18 |
-
],
|
| 19 |
-
"max_position_embeddings": 128,
|
| 20 |
-
"max_window_layers": 28,
|
| 21 |
-
"model_type": "qwen3",
|
| 22 |
-
"num_attention_heads": 3,
|
| 23 |
-
"num_hidden_layers": 2,
|
| 24 |
-
"num_key_value_heads": 3,
|
| 25 |
-
"pad_token_id": null,
|
| 26 |
-
"rms_norm_eps": 1e-06,
|
| 27 |
-
"rope_parameters": {
|
| 28 |
-
"rope_theta": 10000.0,
|
| 29 |
-
"rope_type": "default"
|
| 30 |
-
},
|
| 31 |
-
"sliding_window": null,
|
| 32 |
-
"tie_word_embeddings": false,
|
| 33 |
-
"transformers_version": "5.5.0",
|
| 34 |
-
"use_cache": true,
|
| 35 |
-
"use_sliding_window": false,
|
| 36 |
-
"vocab_size": 151645
|
| 37 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
add_sub_baseline_10K/generation_config.json
DELETED
|
@@ -1,7 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"_from_model_config": true,
|
| 3 |
-
"output_attentions": false,
|
| 4 |
-
"output_hidden_states": false,
|
| 5 |
-
"transformers_version": "5.5.0",
|
| 6 |
-
"use_cache": true
|
| 7 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
add_sub_baseline_10K/metrics.json
DELETED
|
@@ -1,831 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"history": {
|
| 3 |
-
"step": [
|
| 4 |
-
50,
|
| 5 |
-
100,
|
| 6 |
-
150,
|
| 7 |
-
200,
|
| 8 |
-
250,
|
| 9 |
-
300,
|
| 10 |
-
350,
|
| 11 |
-
400,
|
| 12 |
-
450,
|
| 13 |
-
500,
|
| 14 |
-
550,
|
| 15 |
-
600,
|
| 16 |
-
650,
|
| 17 |
-
700,
|
| 18 |
-
750,
|
| 19 |
-
800,
|
| 20 |
-
850,
|
| 21 |
-
900,
|
| 22 |
-
950,
|
| 23 |
-
1000,
|
| 24 |
-
1050,
|
| 25 |
-
1100,
|
| 26 |
-
1150,
|
| 27 |
-
1200,
|
| 28 |
-
1250,
|
| 29 |
-
1300,
|
| 30 |
-
1350,
|
| 31 |
-
1400,
|
| 32 |
-
1450,
|
| 33 |
-
1500,
|
| 34 |
-
1550,
|
| 35 |
-
1600,
|
| 36 |
-
1650,
|
| 37 |
-
1700,
|
| 38 |
-
1750,
|
| 39 |
-
1800,
|
| 40 |
-
1850,
|
| 41 |
-
1900,
|
| 42 |
-
1950,
|
| 43 |
-
2000,
|
| 44 |
-
2050,
|
| 45 |
-
2100,
|
| 46 |
-
2150,
|
| 47 |
-
2200,
|
| 48 |
-
2250,
|
| 49 |
-
2300,
|
| 50 |
-
2350,
|
| 51 |
-
2400,
|
| 52 |
-
2450,
|
| 53 |
-
2500,
|
| 54 |
-
2550,
|
| 55 |
-
2600,
|
| 56 |
-
2650,
|
| 57 |
-
2700,
|
| 58 |
-
2750,
|
| 59 |
-
2800,
|
| 60 |
-
2850,
|
| 61 |
-
2900,
|
| 62 |
-
2950,
|
| 63 |
-
3000,
|
| 64 |
-
3050,
|
| 65 |
-
3100
|
| 66 |
-
],
|
| 67 |
-
"loss": [
|
| 68 |
-
10.006766319274902,
|
| 69 |
-
7.251689434051514,
|
| 70 |
-
6.329094886779785,
|
| 71 |
-
5.457243919372559,
|
| 72 |
-
4.436343669891357,
|
| 73 |
-
3.0709240436553955,
|
| 74 |
-
2.1984894275665283,
|
| 75 |
-
1.9786336421966553,
|
| 76 |
-
1.8310412168502808,
|
| 77 |
-
1.7493754625320435,
|
| 78 |
-
1.7336280345916748,
|
| 79 |
-
1.668373942375183,
|
| 80 |
-
1.656901478767395,
|
| 81 |
-
1.5222007036209106,
|
| 82 |
-
1.1166151762008667,
|
| 83 |
-
0.8200947642326355,
|
| 84 |
-
0.5962659120559692,
|
| 85 |
-
0.4486698508262634,
|
| 86 |
-
0.3298042118549347,
|
| 87 |
-
0.21894291043281555,
|
| 88 |
-
0.1570301353931427,
|
| 89 |
-
0.12834718823432922,
|
| 90 |
-
0.09281289577484131,
|
| 91 |
-
0.0865481048822403,
|
| 92 |
-
0.10090114921331406,
|
| 93 |
-
0.05000796541571617,
|
| 94 |
-
0.05903910472989082,
|
| 95 |
-
0.041255030781030655,
|
| 96 |
-
0.03712769225239754,
|
| 97 |
-
0.05814942717552185,
|
| 98 |
-
0.04444698244333267,
|
| 99 |
-
0.03615949675440788,
|
| 100 |
-
0.027622858062386513,
|
| 101 |
-
0.04681888222694397,
|
| 102 |
-
0.06002098321914673,
|
| 103 |
-
0.049318160861730576,
|
| 104 |
-
0.059921007603406906,
|
| 105 |
-
0.032836705446243286,
|
| 106 |
-
0.061555344611406326,
|
| 107 |
-
0.03171056881546974,
|
| 108 |
-
0.04210656136274338,
|
| 109 |
-
0.04413192719221115,
|
| 110 |
-
0.047493066638708115,
|
| 111 |
-
0.050301071256399155,
|
| 112 |
-
0.02659599296748638,
|
| 113 |
-
0.022891702130436897,
|
| 114 |
-
0.05459100008010864,
|
| 115 |
-
0.017184091731905937,
|
| 116 |
-
0.013001530431210995,
|
| 117 |
-
0.03498239442706108,
|
| 118 |
-
0.02405695430934429,
|
| 119 |
-
0.03834375739097595,
|
| 120 |
-
0.03403574228286743,
|
| 121 |
-
0.014610671438276768,
|
| 122 |
-
0.044524967670440674,
|
| 123 |
-
0.017222251743078232,
|
| 124 |
-
0.03136735409498215,
|
| 125 |
-
0.02747531048953533,
|
| 126 |
-
0.030222097411751747,
|
| 127 |
-
0.02000907063484192,
|
| 128 |
-
0.02265077270567417,
|
| 129 |
-
0.026362445205450058
|
| 130 |
-
],
|
| 131 |
-
"base_loss": [
|
| 132 |
-
10.006766319274902,
|
| 133 |
-
7.251689434051514,
|
| 134 |
-
6.329094886779785,
|
| 135 |
-
5.457243919372559,
|
| 136 |
-
4.436343669891357,
|
| 137 |
-
3.0709240436553955,
|
| 138 |
-
2.1984894275665283,
|
| 139 |
-
1.9786336421966553,
|
| 140 |
-
1.8310412168502808,
|
| 141 |
-
1.7493754625320435,
|
| 142 |
-
1.7336280345916748,
|
| 143 |
-
1.668373942375183,
|
| 144 |
-
1.656901478767395,
|
| 145 |
-
1.5222007036209106,
|
| 146 |
-
1.1166151762008667,
|
| 147 |
-
0.8200947642326355,
|
| 148 |
-
0.5962659120559692,
|
| 149 |
-
0.4486698508262634,
|
| 150 |
-
0.3298042118549347,
|
| 151 |
-
0.21894291043281555,
|
| 152 |
-
0.1570301353931427,
|
| 153 |
-
0.12834718823432922,
|
| 154 |
-
0.09281289577484131,
|
| 155 |
-
0.0865481048822403,
|
| 156 |
-
0.10090114921331406,
|
| 157 |
-
0.05000796541571617,
|
| 158 |
-
0.05903910472989082,
|
| 159 |
-
0.041255030781030655,
|
| 160 |
-
0.03712769225239754,
|
| 161 |
-
0.05814942717552185,
|
| 162 |
-
0.04444698244333267,
|
| 163 |
-
0.03615949675440788,
|
| 164 |
-
0.027622858062386513,
|
| 165 |
-
0.04681888222694397,
|
| 166 |
-
0.06002098321914673,
|
| 167 |
-
0.049318160861730576,
|
| 168 |
-
0.059921007603406906,
|
| 169 |
-
0.032836705446243286,
|
| 170 |
-
0.061555344611406326,
|
| 171 |
-
0.03171056881546974,
|
| 172 |
-
0.04210656136274338,
|
| 173 |
-
0.04413192719221115,
|
| 174 |
-
0.047493066638708115,
|
| 175 |
-
0.050301071256399155,
|
| 176 |
-
0.02659599296748638,
|
| 177 |
-
0.022891702130436897,
|
| 178 |
-
0.05459100008010864,
|
| 179 |
-
0.017184091731905937,
|
| 180 |
-
0.013001530431210995,
|
| 181 |
-
0.03498239442706108,
|
| 182 |
-
0.02405695430934429,
|
| 183 |
-
0.03834375739097595,
|
| 184 |
-
0.03403574228286743,
|
| 185 |
-
0.014610671438276768,
|
| 186 |
-
0.044524967670440674,
|
| 187 |
-
0.017222251743078232,
|
| 188 |
-
0.03136735409498215,
|
| 189 |
-
0.02747531048953533,
|
| 190 |
-
0.030222097411751747,
|
| 191 |
-
0.02000907063484192,
|
| 192 |
-
0.02265077270567417,
|
| 193 |
-
0.026362445205450058
|
| 194 |
-
],
|
| 195 |
-
"lr": [
|
| 196 |
-
6.242038216560511e-06,
|
| 197 |
-
1.2611464968152866e-05,
|
| 198 |
-
1.8980891719745225e-05,
|
| 199 |
-
2.5350318471337578e-05,
|
| 200 |
-
3.1719745222929934e-05,
|
| 201 |
-
3.808917197452229e-05,
|
| 202 |
-
4.445859872611465e-05,
|
| 203 |
-
5.082802547770701e-05,
|
| 204 |
-
5.7197452229299365e-05,
|
| 205 |
-
6.356687898089173e-05,
|
| 206 |
-
6.993630573248408e-05,
|
| 207 |
-
7.630573248407644e-05,
|
| 208 |
-
7.99862055592881e-05,
|
| 209 |
-
7.984241248831029e-05,
|
| 210 |
-
7.954287783192742e-05,
|
| 211 |
-
7.90887724530305e-05,
|
| 212 |
-
7.848187142213441e-05,
|
| 213 |
-
7.772454707873448e-05,
|
| 214 |
-
7.681975975797462e-05,
|
| 215 |
-
7.57710462188759e-05,
|
| 216 |
-
7.458250581935905e-05,
|
| 217 |
-
7.325878449210182e-05,
|
| 218 |
-
7.180505658386849e-05,
|
| 219 |
-
7.022700462930083e-05,
|
| 220 |
-
6.853079713823312e-05,
|
| 221 |
-
6.672306448335957e-05,
|
| 222 |
-
6.481087298250779e-05,
|
| 223 |
-
6.280169727682872e-05,
|
| 224 |
-
6.070339111287581e-05,
|
| 225 |
-
5.8524156642783655e-05,
|
| 226 |
-
5.627251236255051e-05,
|
| 227 |
-
5.3957259813751526e-05,
|
| 228 |
-
5.1587449178844164e-05,
|
| 229 |
-
4.917234390455111e-05,
|
| 230 |
-
4.672138449160635e-05,
|
| 231 |
-
4.424415159240753e-05,
|
| 232 |
-
4.175032856082417e-05,
|
| 233 |
-
3.924966360055181e-05,
|
| 234 |
-
3.675193165997228e-05,
|
| 235 |
-
3.42668962224704e-05,
|
| 236 |
-
3.180427114156694e-05,
|
| 237 |
-
2.9373682670051437e-05,
|
| 238 |
-
2.6984631831541183e-05,
|
| 239 |
-
2.4646457281553407e-05,
|
| 240 |
-
2.2368298803264487e-05,
|
| 241 |
-
2.0159061580649347e-05,
|
| 242 |
-
1.8027381388654794e-05,
|
| 243 |
-
1.5981590836476463e-05,
|
| 244 |
-
1.4029686795892575e-05,
|
| 245 |
-
1.2179299141974771e-05,
|
| 246 |
-
1.0437660928367057e-05,
|
| 247 |
-
8.811580113715755e-06,
|
| 248 |
-
7.307412949770034e-06,
|
| 249 |
-
5.9310391351775455e-06,
|
| 250 |
-
4.687838832097362e-06,
|
| 251 |
-
3.5826716354707645e-06,
|
| 252 |
-
2.6198575771580583e-06,
|
| 253 |
-
1.8031602391947344e-06,
|
| 254 |
-
1.1357720421765062e-06,
|
| 255 |
-
6.203017662798872e-07,
|
| 256 |
-
2.5876435369797334e-07,
|
| 257 |
-
5.257303235302935e-08
|
| 258 |
-
],
|
| 259 |
-
"eval_step": [
|
| 260 |
-
156,
|
| 261 |
-
312,
|
| 262 |
-
468,
|
| 263 |
-
624,
|
| 264 |
-
780,
|
| 265 |
-
936,
|
| 266 |
-
1092,
|
| 267 |
-
1248,
|
| 268 |
-
1404,
|
| 269 |
-
1560,
|
| 270 |
-
1716,
|
| 271 |
-
1872,
|
| 272 |
-
2028,
|
| 273 |
-
2184,
|
| 274 |
-
2340,
|
| 275 |
-
2496,
|
| 276 |
-
2652,
|
| 277 |
-
2808,
|
| 278 |
-
2964,
|
| 279 |
-
3120
|
| 280 |
-
],
|
| 281 |
-
"eval_epoch": [
|
| 282 |
-
1,
|
| 283 |
-
2,
|
| 284 |
-
3,
|
| 285 |
-
4,
|
| 286 |
-
5,
|
| 287 |
-
6,
|
| 288 |
-
7,
|
| 289 |
-
8,
|
| 290 |
-
9,
|
| 291 |
-
10,
|
| 292 |
-
11,
|
| 293 |
-
12,
|
| 294 |
-
13,
|
| 295 |
-
14,
|
| 296 |
-
15,
|
| 297 |
-
16,
|
| 298 |
-
17,
|
| 299 |
-
18,
|
| 300 |
-
19,
|
| 301 |
-
20
|
| 302 |
-
],
|
| 303 |
-
"eval_accuracy": [
|
| 304 |
-
0.0014285714285714286,
|
| 305 |
-
0.012142857142857143,
|
| 306 |
-
0.004285714285714286,
|
| 307 |
-
0.007142857142857143,
|
| 308 |
-
0.035,
|
| 309 |
-
0.26285714285714284,
|
| 310 |
-
0.6278571428571429,
|
| 311 |
-
0.6864285714285714,
|
| 312 |
-
0.6885714285714286,
|
| 313 |
-
0.7314285714285714,
|
| 314 |
-
0.7442857142857143,
|
| 315 |
-
0.73,
|
| 316 |
-
0.735,
|
| 317 |
-
0.75,
|
| 318 |
-
0.7807142857142857,
|
| 319 |
-
0.7964285714285714,
|
| 320 |
-
0.8185714285714286,
|
| 321 |
-
0.8185714285714286,
|
| 322 |
-
0.8221428571428572,
|
| 323 |
-
0.8235714285714286
|
| 324 |
-
]
|
| 325 |
-
},
|
| 326 |
-
"final_accuracy": 0.7609259259259259,
|
| 327 |
-
"sft_eval": {
|
| 328 |
-
"config": {
|
| 329 |
-
"ops": "add_sub",
|
| 330 |
-
"K": null,
|
| 331 |
-
"mode": "sft",
|
| 332 |
-
"n_digits": 6,
|
| 333 |
-
"n_per_split": 250
|
| 334 |
-
},
|
| 335 |
-
"splits": {
|
| 336 |
-
"add_S0": {
|
| 337 |
-
"full_accuracy": 0.988,
|
| 338 |
-
"n_examples": 250,
|
| 339 |
-
"per_subtask": {
|
| 340 |
-
"SA": {
|
| 341 |
-
"accuracy": 0.9980302035456337,
|
| 342 |
-
"count": 1523
|
| 343 |
-
},
|
| 344 |
-
"SS": {
|
| 345 |
-
"accuracy": 0.9955947136563876,
|
| 346 |
-
"count": 227
|
| 347 |
-
}
|
| 348 |
-
}
|
| 349 |
-
},
|
| 350 |
-
"add_S1": {
|
| 351 |
-
"full_accuracy": 0.996,
|
| 352 |
-
"n_examples": 250,
|
| 353 |
-
"per_subtask": {
|
| 354 |
-
"SA": {
|
| 355 |
-
"accuracy": 1.0,
|
| 356 |
-
"count": 542
|
| 357 |
-
},
|
| 358 |
-
"SC": {
|
| 359 |
-
"accuracy": 0.9976019184652278,
|
| 360 |
-
"count": 417
|
| 361 |
-
},
|
| 362 |
-
"SS": {
|
| 363 |
-
"accuracy": 1.0,
|
| 364 |
-
"count": 70
|
| 365 |
-
},
|
| 366 |
-
"UC": {
|
| 367 |
-
"accuracy": 1.0,
|
| 368 |
-
"count": 721
|
| 369 |
-
}
|
| 370 |
-
}
|
| 371 |
-
},
|
| 372 |
-
"add_S2": {
|
| 373 |
-
"full_accuracy": 0.968,
|
| 374 |
-
"n_examples": 250,
|
| 375 |
-
"per_subtask": {
|
| 376 |
-
"SA": {
|
| 377 |
-
"accuracy": 1.0,
|
| 378 |
-
"count": 368
|
| 379 |
-
},
|
| 380 |
-
"SC": {
|
| 381 |
-
"accuracy": 0.9750778816199377,
|
| 382 |
-
"count": 321
|
| 383 |
-
},
|
| 384 |
-
"SS": {
|
| 385 |
-
"accuracy": 1.0,
|
| 386 |
-
"count": 228
|
| 387 |
-
},
|
| 388 |
-
"UC": {
|
| 389 |
-
"accuracy": 1.0,
|
| 390 |
-
"count": 531
|
| 391 |
-
},
|
| 392 |
-
"US": {
|
| 393 |
-
"accuracy": 1.0,
|
| 394 |
-
"count": 302
|
| 395 |
-
}
|
| 396 |
-
}
|
| 397 |
-
},
|
| 398 |
-
"add_S3": {
|
| 399 |
-
"full_accuracy": 0.74,
|
| 400 |
-
"n_examples": 250,
|
| 401 |
-
"per_subtask": {
|
| 402 |
-
"SA": {
|
| 403 |
-
"accuracy": 1.0,
|
| 404 |
-
"count": 307
|
| 405 |
-
},
|
| 406 |
-
"SC": {
|
| 407 |
-
"accuracy": 0.9965635738831615,
|
| 408 |
-
"count": 291
|
| 409 |
-
},
|
| 410 |
-
"SS": {
|
| 411 |
-
"accuracy": 1.0,
|
| 412 |
-
"count": 113
|
| 413 |
-
},
|
| 414 |
-
"UC": {
|
| 415 |
-
"accuracy": 0.8674948240165632,
|
| 416 |
-
"count": 483
|
| 417 |
-
},
|
| 418 |
-
"US": {
|
| 419 |
-
"accuracy": 1.0,
|
| 420 |
-
"count": 556
|
| 421 |
-
}
|
| 422 |
-
}
|
| 423 |
-
},
|
| 424 |
-
"add_S4": {
|
| 425 |
-
"full_accuracy": 0.704,
|
| 426 |
-
"n_examples": 250,
|
| 427 |
-
"per_subtask": {
|
| 428 |
-
"SA": {
|
| 429 |
-
"accuracy": 1.0,
|
| 430 |
-
"count": 238
|
| 431 |
-
},
|
| 432 |
-
"SC": {
|
| 433 |
-
"accuracy": 1.0,
|
| 434 |
-
"count": 271
|
| 435 |
-
},
|
| 436 |
-
"SS": {
|
| 437 |
-
"accuracy": 1.0,
|
| 438 |
-
"count": 59
|
| 439 |
-
},
|
| 440 |
-
"UC": {
|
| 441 |
-
"accuracy": 0.8395061728395061,
|
| 442 |
-
"count": 405
|
| 443 |
-
},
|
| 444 |
-
"US": {
|
| 445 |
-
"accuracy": 0.9317889317889317,
|
| 446 |
-
"count": 777
|
| 447 |
-
}
|
| 448 |
-
}
|
| 449 |
-
},
|
| 450 |
-
"add_S5": {
|
| 451 |
-
"full_accuracy": 0.5,
|
| 452 |
-
"n_examples": 250,
|
| 453 |
-
"per_subtask": {
|
| 454 |
-
"SA": {
|
| 455 |
-
"accuracy": 1.0,
|
| 456 |
-
"count": 250
|
| 457 |
-
},
|
| 458 |
-
"SC": {
|
| 459 |
-
"accuracy": 1.0,
|
| 460 |
-
"count": 250
|
| 461 |
-
},
|
| 462 |
-
"UC": {
|
| 463 |
-
"accuracy": 0.644,
|
| 464 |
-
"count": 250
|
| 465 |
-
},
|
| 466 |
-
"US": {
|
| 467 |
-
"accuracy": 0.795,
|
| 468 |
-
"count": 1000
|
| 469 |
-
}
|
| 470 |
-
}
|
| 471 |
-
},
|
| 472 |
-
"add_S6": {
|
| 473 |
-
"full_accuracy": 0.916,
|
| 474 |
-
"n_examples": 250,
|
| 475 |
-
"per_subtask": {
|
| 476 |
-
"SC": {
|
| 477 |
-
"accuracy": 1.0,
|
| 478 |
-
"count": 250
|
| 479 |
-
},
|
| 480 |
-
"UC": {
|
| 481 |
-
"accuracy": 1.0,
|
| 482 |
-
"count": 250
|
| 483 |
-
},
|
| 484 |
-
"US": {
|
| 485 |
-
"accuracy": 0.9824,
|
| 486 |
-
"count": 1250
|
| 487 |
-
}
|
| 488 |
-
}
|
| 489 |
-
},
|
| 490 |
-
"add_random": {
|
| 491 |
-
"full_accuracy": 0.985,
|
| 492 |
-
"n_examples": 200,
|
| 493 |
-
"per_subtask": {
|
| 494 |
-
"SA": {
|
| 495 |
-
"accuracy": 0.9977324263038548,
|
| 496 |
-
"count": 441
|
| 497 |
-
},
|
| 498 |
-
"SC": {
|
| 499 |
-
"accuracy": 1.0,
|
| 500 |
-
"count": 317
|
| 501 |
-
},
|
| 502 |
-
"SS": {
|
| 503 |
-
"accuracy": 1.0,
|
| 504 |
-
"count": 54
|
| 505 |
-
},
|
| 506 |
-
"UC": {
|
| 507 |
-
"accuracy": 0.9962406015037594,
|
| 508 |
-
"count": 532
|
| 509 |
-
},
|
| 510 |
-
"US": {
|
| 511 |
-
"accuracy": 0.9821428571428571,
|
| 512 |
-
"count": 56
|
| 513 |
-
}
|
| 514 |
-
}
|
| 515 |
-
},
|
| 516 |
-
"add_C3": {
|
| 517 |
-
"full_accuracy": 0.812,
|
| 518 |
-
"n_examples": 250,
|
| 519 |
-
"per_subtask": {
|
| 520 |
-
"SA": {
|
| 521 |
-
"accuracy": 1.0,
|
| 522 |
-
"count": 750
|
| 523 |
-
},
|
| 524 |
-
"SC": {
|
| 525 |
-
"accuracy": 1.0,
|
| 526 |
-
"count": 250
|
| 527 |
-
},
|
| 528 |
-
"UC": {
|
| 529 |
-
"accuracy": 0.9012605042016807,
|
| 530 |
-
"count": 476
|
| 531 |
-
},
|
| 532 |
-
"US": {
|
| 533 |
-
"accuracy": 1.0,
|
| 534 |
-
"count": 274
|
| 535 |
-
}
|
| 536 |
-
}
|
| 537 |
-
},
|
| 538 |
-
"add_C4": {
|
| 539 |
-
"full_accuracy": 0.888,
|
| 540 |
-
"n_examples": 250,
|
| 541 |
-
"per_subtask": {
|
| 542 |
-
"SA": {
|
| 543 |
-
"accuracy": 1.0,
|
| 544 |
-
"count": 500
|
| 545 |
-
},
|
| 546 |
-
"SC": {
|
| 547 |
-
"accuracy": 1.0,
|
| 548 |
-
"count": 250
|
| 549 |
-
},
|
| 550 |
-
"UC": {
|
| 551 |
-
"accuracy": 0.9606656580937972,
|
| 552 |
-
"count": 661
|
| 553 |
-
},
|
| 554 |
-
"US": {
|
| 555 |
-
"accuracy": 0.967551622418879,
|
| 556 |
-
"count": 339
|
| 557 |
-
}
|
| 558 |
-
}
|
| 559 |
-
},
|
| 560 |
-
"add_C5": {
|
| 561 |
-
"full_accuracy": 0.812,
|
| 562 |
-
"n_examples": 250,
|
| 563 |
-
"per_subtask": {
|
| 564 |
-
"SA": {
|
| 565 |
-
"accuracy": 1.0,
|
| 566 |
-
"count": 250
|
| 567 |
-
},
|
| 568 |
-
"SC": {
|
| 569 |
-
"accuracy": 1.0,
|
| 570 |
-
"count": 250
|
| 571 |
-
},
|
| 572 |
-
"UC": {
|
| 573 |
-
"accuracy": 0.9385245901639344,
|
| 574 |
-
"count": 732
|
| 575 |
-
},
|
| 576 |
-
"US": {
|
| 577 |
-
"accuracy": 0.9498069498069498,
|
| 578 |
-
"count": 518
|
| 579 |
-
}
|
| 580 |
-
}
|
| 581 |
-
},
|
| 582 |
-
"add_C6": {
|
| 583 |
-
"full_accuracy": 0.824,
|
| 584 |
-
"n_examples": 250,
|
| 585 |
-
"per_subtask": {
|
| 586 |
-
"SC": {
|
| 587 |
-
"accuracy": 1.0,
|
| 588 |
-
"count": 250
|
| 589 |
-
},
|
| 590 |
-
"UC": {
|
| 591 |
-
"accuracy": 0.954954954954955,
|
| 592 |
-
"count": 888
|
| 593 |
-
},
|
| 594 |
-
"US": {
|
| 595 |
-
"accuracy": 0.9624183006535948,
|
| 596 |
-
"count": 612
|
| 597 |
-
}
|
| 598 |
-
}
|
| 599 |
-
},
|
| 600 |
-
"sub_M0": {
|
| 601 |
-
"full_accuracy": 0.968,
|
| 602 |
-
"n_examples": 250,
|
| 603 |
-
"per_subtask": {
|
| 604 |
-
"MD": {
|
| 605 |
-
"accuracy": 0.9946843853820598,
|
| 606 |
-
"count": 1505
|
| 607 |
-
},
|
| 608 |
-
"ME": {
|
| 609 |
-
"accuracy": 1.0,
|
| 610 |
-
"count": 245
|
| 611 |
-
}
|
| 612 |
-
}
|
| 613 |
-
},
|
| 614 |
-
"sub_M1": {
|
| 615 |
-
"full_accuracy": 1.0,
|
| 616 |
-
"n_examples": 250,
|
| 617 |
-
"per_subtask": {
|
| 618 |
-
"MD": {
|
| 619 |
-
"accuracy": 1.0,
|
| 620 |
-
"count": 714
|
| 621 |
-
},
|
| 622 |
-
"MB": {
|
| 623 |
-
"accuracy": 1.0,
|
| 624 |
-
"count": 374
|
| 625 |
-
},
|
| 626 |
-
"ME": {
|
| 627 |
-
"accuracy": 1.0,
|
| 628 |
-
"count": 75
|
| 629 |
-
},
|
| 630 |
-
"UB": {
|
| 631 |
-
"accuracy": 1.0,
|
| 632 |
-
"count": 587
|
| 633 |
-
}
|
| 634 |
-
}
|
| 635 |
-
},
|
| 636 |
-
"sub_M2": {
|
| 637 |
-
"full_accuracy": 0.996,
|
| 638 |
-
"n_examples": 250,
|
| 639 |
-
"per_subtask": {
|
| 640 |
-
"MD": {
|
| 641 |
-
"accuracy": 0.9981949458483754,
|
| 642 |
-
"count": 554
|
| 643 |
-
},
|
| 644 |
-
"MB": {
|
| 645 |
-
"accuracy": 1.0,
|
| 646 |
-
"count": 273
|
| 647 |
-
},
|
| 648 |
-
"ME": {
|
| 649 |
-
"accuracy": 1.0,
|
| 650 |
-
"count": 219
|
| 651 |
-
},
|
| 652 |
-
"UB": {
|
| 653 |
-
"accuracy": 1.0,
|
| 654 |
-
"count": 430
|
| 655 |
-
},
|
| 656 |
-
"UD": {
|
| 657 |
-
"accuracy": 1.0,
|
| 658 |
-
"count": 274
|
| 659 |
-
}
|
| 660 |
-
}
|
| 661 |
-
},
|
| 662 |
-
"sub_M3": {
|
| 663 |
-
"full_accuracy": 0.34,
|
| 664 |
-
"n_examples": 250,
|
| 665 |
-
"per_subtask": {
|
| 666 |
-
"MD": {
|
| 667 |
-
"accuracy": 1.0,
|
| 668 |
-
"count": 458
|
| 669 |
-
},
|
| 670 |
-
"MB": {
|
| 671 |
-
"accuracy": 1.0,
|
| 672 |
-
"count": 261
|
| 673 |
-
},
|
| 674 |
-
"ME": {
|
| 675 |
-
"accuracy": 1.0,
|
| 676 |
-
"count": 124
|
| 677 |
-
},
|
| 678 |
-
"UB": {
|
| 679 |
-
"accuracy": 0.5747422680412371,
|
| 680 |
-
"count": 388
|
| 681 |
-
},
|
| 682 |
-
"UD": {
|
| 683 |
-
"accuracy": 1.0,
|
| 684 |
-
"count": 519
|
| 685 |
-
}
|
| 686 |
-
}
|
| 687 |
-
},
|
| 688 |
-
"sub_M4": {
|
| 689 |
-
"full_accuracy": 0.116,
|
| 690 |
-
"n_examples": 250,
|
| 691 |
-
"per_subtask": {
|
| 692 |
-
"MD": {
|
| 693 |
-
"accuracy": 1.0,
|
| 694 |
-
"count": 500
|
| 695 |
-
},
|
| 696 |
-
"MB": {
|
| 697 |
-
"accuracy": 1.0,
|
| 698 |
-
"count": 250
|
| 699 |
-
},
|
| 700 |
-
"UB": {
|
| 701 |
-
"accuracy": 0.312,
|
| 702 |
-
"count": 250
|
| 703 |
-
},
|
| 704 |
-
"UD": {
|
| 705 |
-
"accuracy": 0.7466666666666667,
|
| 706 |
-
"count": 750
|
| 707 |
-
}
|
| 708 |
-
}
|
| 709 |
-
},
|
| 710 |
-
"sub_M5": {
|
| 711 |
-
"full_accuracy": 0.052,
|
| 712 |
-
"n_examples": 250,
|
| 713 |
-
"per_subtask": {
|
| 714 |
-
"MD": {
|
| 715 |
-
"accuracy": 1.0,
|
| 716 |
-
"count": 250
|
| 717 |
-
},
|
| 718 |
-
"MB": {
|
| 719 |
-
"accuracy": 1.0,
|
| 720 |
-
"count": 250
|
| 721 |
-
},
|
| 722 |
-
"UB": {
|
| 723 |
-
"accuracy": 0.368,
|
| 724 |
-
"count": 250
|
| 725 |
-
},
|
| 726 |
-
"UD": {
|
| 727 |
-
"accuracy": 0.547,
|
| 728 |
-
"count": 1000
|
| 729 |
-
}
|
| 730 |
-
}
|
| 731 |
-
},
|
| 732 |
-
"sub_random": {
|
| 733 |
-
"full_accuracy": 0.99,
|
| 734 |
-
"n_examples": 200,
|
| 735 |
-
"per_subtask": {
|
| 736 |
-
"MD": {
|
| 737 |
-
"accuracy": 1.0,
|
| 738 |
-
"count": 580
|
| 739 |
-
},
|
| 740 |
-
"MB": {
|
| 741 |
-
"accuracy": 1.0,
|
| 742 |
-
"count": 267
|
| 743 |
-
},
|
| 744 |
-
"ME": {
|
| 745 |
-
"accuracy": 1.0,
|
| 746 |
-
"count": 63
|
| 747 |
-
},
|
| 748 |
-
"UB": {
|
| 749 |
-
"accuracy": 0.9955357142857143,
|
| 750 |
-
"count": 448
|
| 751 |
-
},
|
| 752 |
-
"UD": {
|
| 753 |
-
"accuracy": 1.0,
|
| 754 |
-
"count": 42
|
| 755 |
-
}
|
| 756 |
-
}
|
| 757 |
-
},
|
| 758 |
-
"sub_B3": {
|
| 759 |
-
"full_accuracy": 0.868,
|
| 760 |
-
"n_examples": 250,
|
| 761 |
-
"per_subtask": {
|
| 762 |
-
"MD": {
|
| 763 |
-
"accuracy": 1.0,
|
| 764 |
-
"count": 750
|
| 765 |
-
},
|
| 766 |
-
"MB": {
|
| 767 |
-
"accuracy": 1.0,
|
| 768 |
-
"count": 250
|
| 769 |
-
},
|
| 770 |
-
"UB": {
|
| 771 |
-
"accuracy": 0.9350393700787402,
|
| 772 |
-
"count": 508
|
| 773 |
-
},
|
| 774 |
-
"UD": {
|
| 775 |
-
"accuracy": 1.0,
|
| 776 |
-
"count": 242
|
| 777 |
-
}
|
| 778 |
-
}
|
| 779 |
-
},
|
| 780 |
-
"sub_B4": {
|
| 781 |
-
"full_accuracy": 0.716,
|
| 782 |
-
"n_examples": 250,
|
| 783 |
-
"per_subtask": {
|
| 784 |
-
"MD": {
|
| 785 |
-
"accuracy": 1.0,
|
| 786 |
-
"count": 500
|
| 787 |
-
},
|
| 788 |
-
"MB": {
|
| 789 |
-
"accuracy": 1.0,
|
| 790 |
-
"count": 250
|
| 791 |
-
},
|
| 792 |
-
"UB": {
|
| 793 |
-
"accuracy": 0.8973941368078175,
|
| 794 |
-
"count": 614
|
| 795 |
-
},
|
| 796 |
-
"UD": {
|
| 797 |
-
"accuracy": 0.9300518134715026,
|
| 798 |
-
"count": 386
|
| 799 |
-
}
|
| 800 |
-
}
|
| 801 |
-
},
|
| 802 |
-
"sub_B5": {
|
| 803 |
-
"full_accuracy": 0.652,
|
| 804 |
-
"n_examples": 250,
|
| 805 |
-
"per_subtask": {
|
| 806 |
-
"MD": {
|
| 807 |
-
"accuracy": 1.0,
|
| 808 |
-
"count": 250
|
| 809 |
-
},
|
| 810 |
-
"MB": {
|
| 811 |
-
"accuracy": 1.0,
|
| 812 |
-
"count": 250
|
| 813 |
-
},
|
| 814 |
-
"UB": {
|
| 815 |
-
"accuracy": 0.8957528957528957,
|
| 816 |
-
"count": 777
|
| 817 |
-
},
|
| 818 |
-
"UD": {
|
| 819 |
-
"accuracy": 0.9090909090909091,
|
| 820 |
-
"count": 473
|
| 821 |
-
}
|
| 822 |
-
}
|
| 823 |
-
}
|
| 824 |
-
},
|
| 825 |
-
"summary": {
|
| 826 |
-
"overall_accuracy": 0.7609259259259259,
|
| 827 |
-
"total_examples": 5400,
|
| 828 |
-
"n_splits": 22
|
| 829 |
-
}
|
| 830 |
-
}
|
| 831 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
add_sub_baseline_10K/model.safetensors
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:bd67ff426b2d74cb962e75cb59ccc6e89358ef5a3fa76bfe35640b18cc9c69ee
|
| 3 |
-
size 650266922
|
|
|
|
|
|
|
|
|
|
|
|
add_sub_baseline_10K/train_config.json
DELETED
|
@@ -1,35 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"mode": "baseline",
|
| 3 |
-
"ops": "add_sub",
|
| 4 |
-
"n_digits": 6,
|
| 5 |
-
"n_layer": 2,
|
| 6 |
-
"n_head": 3,
|
| 7 |
-
"n_embd": 510,
|
| 8 |
-
"abs_vocab": 0,
|
| 9 |
-
"K": 4,
|
| 10 |
-
"alpha_info_gain": 10.0,
|
| 11 |
-
"alpha_abs": 0.1,
|
| 12 |
-
"alpha_soft_zipf": 1.0,
|
| 13 |
-
"batch_size": 64,
|
| 14 |
-
"num_epochs": 20,
|
| 15 |
-
"dataset_size": 10000,
|
| 16 |
-
"lr": 8e-05,
|
| 17 |
-
"output_dir": "ckpt/sweep/add_sub_baseline_10K",
|
| 18 |
-
"device": "cuda",
|
| 19 |
-
"push_to_hub": true,
|
| 20 |
-
"no_wandb": false,
|
| 21 |
-
"n_params": 162490082,
|
| 22 |
-
"run_name": "add_sub_baseline_10K",
|
| 23 |
-
"git_commit": "800625019270114adcda289bbd550c4f1109a514",
|
| 24 |
-
"timestamp": "2026-04-11T21:33:55.773262+00:00",
|
| 25 |
-
"tokenizer": "Qwen/Qwen3-0.6B",
|
| 26 |
-
"dataset_repo": "thoughtworks/arithmetic-sorl-data",
|
| 27 |
-
"dataset_config": "add_sub_6digit",
|
| 28 |
-
"model_repo": "thoughtworks/arithmetic-sorl",
|
| 29 |
-
"trainer_version": "sft",
|
| 30 |
-
"wandb_run_id": "1e2dttjb",
|
| 31 |
-
"wandb_url": "https://wandb.ai/nlp_and_interpretability/sorl-arithmetic/runs/1e2dttjb",
|
| 32 |
-
"final_accuracy": 0.7609259259259259,
|
| 33 |
-
"sft_accuracy": 0.7609259259259259,
|
| 34 |
-
"eval_method": "ArithmeticEvaluator"
|
| 35 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|