Sample stimuli

sample 0 sample 1 sample 2 sample 3 sample 4 sample 5 sample 6 sample 7 sample 8 sample 9

How to use

from brainscore_vision import load_benchmark
benchmark = load_benchmark("Geirhos2021silhouette-error_consistency")
score = benchmark(my_model)

Model scores

Min Alignment Max Alignment

Rank

Model

Score

1
1.2
2
1.2
3
1.2
4
1.1
5
1.1
6
1.1
7
1.1
8
1.1
9
1.0
10
1.0
11
1.0
12
.993
13
.990
14
.966
15
.966
16
.959
17
.936
18
.914
19
.908
20
.908
21
.904
22
.900
23
.897
24
.891
25
.888
26
.883
27
.870
28
.856
29
.848
30
.841
31
.841
32
.840
33
.827
34
.825
35
.822
36
.818
37
.817
38
.810
39
.793
40
.781
41
.777
42
.773
43
.767
44
.763
45
.763
46
.750
47
.749
48
.737
49
.728
50
.728
51
.718
52
.706
53
.704
54
.701
55
.685
56
.681
57
.675
58
.674
59
.668
60
.663
61
.660
62
.658
63
.658
64
.648
65
.648
66
.647
67
.646
68
.640
69
.628
70
.618
71
.615
72
.614
73
.610
74
.610
75
.608
76
.602
77
.601
78
.584
79
.578
80
.574
81
.560
82
.558
83
.554
84
.553
85
.553
86
.551
87
.549
88
.548
89
.546
90
.546
91
.546
92
.546
93
.544
94
.535
95
.534
96
.531
97
.524
98
.523
99
.521
100
.521
101
.520
102
.520
103
.510
104
.510
105
.509
106
.500
107
.497
108
.492
109
.492
110
.492
111
.492
112
.491
113
.491
114
.491
115
.482
116
.482
117
.482
118
.478
119
.474
120
.474
121
.469
122
.469
123
.468
124
.465
125
.463
126
.461
127
.458
128
.451
129
.450
130
.449
131
.445
132
.439
133
.439
134
.438
135
.437
136
.434
137
.428
138
.421
139
.418
140
.417
141
.409
142
.395
143
.394
144
.392
145
.390
146
.388
147
.384
148
.379
149
.378
150
.378
151
.375
152
.373
153
.369
154
.369
155
.365
156
.364
157
.362
158
.360
159
.358
160
.358
161
.357
162
.353
163
.353
164
.352
165
.349
166
.336
167
.336
168
.335
169
.333
170
.332
171
.328
172
.317
173
.315
174
.314
175
.310
176
.309
177
.304
178
.298
179
.297
180
.296
181
.296
182
.287
183
.287
184
.286
185
.285
186
.280
187
.279
188
.276
189
.270
190
.266
191
.263
192
.262
193
.257
194
.256
195
.255
196
.252
197
.252
198
.249
199
.247
200
.246
201
.245
202
.243
203
.240
204
.236
205
.236
206
.231
207
.231
208
.228
209
.220
210
.214
211
.214
212
.205
213
.204
214
.204
215
.203
216
.203
217
.203
218
.203
219
.203
220
.203
221
.203
222
.203
223
.203
224
.203
225
.203
226
.203
227
.203
228
.194
229
.194
230
.193
231
.192
232
.186
233
.185
234
.185
235
.185
236
.180
237
.176
238
.169
239
.166
240
.164
241
.163
242
.161
243
.159
244
.154
245
.150
246
.149
247
.149
248
.149
249
.148
250
.147
251
.146
252
.145
253
.143
254
.139
255
.139
256
.139
257
.139
258
.139
259
.139
260
.139
261
.139
262
.135
263
.134
264
.129
265
.127
266
.118
267
.117
268
.117
269
.112
270
.109
271
.107
272
.106
273
.102
274
.102
275
.098
276
.090
277
.089
278
.089
279
.084
280
.084
281
.084
282
.083
283
.082
284
.082
285
.081
286
.077
287
.077
288
.077
289
.077
290
.077
291
.075
292
.073
293
.070
294
.066
295
.065
296
.059
297
.056
298
.054
299
.053
300
.052
301
.051
302
.050
303
.049
304
.046
305
.043
306
.042
307
.041
308
.033
309
.023
310
.023
311
.012
312
.004
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460

Benchmark bibtex

@article{geirhos2021partial,
              title={Partial success in closing the gap between human and machine vision},
              author={Geirhos, Robert and Narayanappa, Kantharaju and Mitzkus, Benjamin and Thieringer, Tizian and Bethge, Matthias and Wichmann, Felix A and Brendel, Wieland},
              journal={Advances in Neural Information Processing Systems},
              volume={34},
              year={2021},
              url={https://openreview.net/forum?id=QkljT4mrfs}
        }

Ceiling

0.48.

Note that scores are relative to this ceiling.

Data: Geirhos2021silhouette

Metric: error_consistency