Brain-Score - Blank2014-ridge

Leaderboard Explore Compare Profile Community Blog FAQ Competition 2024 Report Issue

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_language import load_benchmark
benchmark = load_benchmark("Blank2014-ridge")
score = benchmark(my_model)

Model scores

Score Legend Min Alignment Max Alignment

Rank	Model	Score
1	mistral-7b	.259
2	distilgpt2	.176
3	tinyllama-1.1b	.142
4	gpt2	.121
5	falcon-7b	.109
6	opt-2.7b	.104
7	gpt-neo-1.3B	.104
8	qwen2.5-3b	.100
9	gpt-neo-2.7B	.096
10	openai-gpt	.093
11	opt-6.7b	.054
12	randomembedding-100	.045
13	gpt2-large	.041
14	gpt2-medium	.025
15	oasm-sigma0.4	.000
16	oasm-sigma0.5	.000
17	oasm-sigma0.6	.000
18	oasm-sigma0.3	.000
19	oasm-sigma0	.000
20	gpt-neo-125m	.000
21	gpt2-xl	.000
22	glove-840b	.000
23	randomembedding-1600	.000
24	earley-parser-minivocab
25	lm1b
26	phi-2
27	gemma-2b

Benchmark bibtex

None

Ceiling

0.16.

Note that scores are relative to this ceiling.

Data: Blank2014

Metric: ridge