Llama 3.1 8B Instruct Behavioral Fingerprint

How Meta's Llama 3.1 8B Instruct answers one-word probe questions at temperature 1.0 — random numbers, colors, animals, coin flips — measured over 434 valid samples in the open dataset of "One Token Is Enough" (arXiv:2607.10252).

Vendor

Meta

Data generated

2026-07-08

Probe cells

15

Valid samples

434

Randomness score

55 / 100

Mean normalized entropy across cells, 0–100. 100 would be a uniform-random baseline; real models score far lower.

Signature answers

Favorite random number (1–100)

53

6 / 30 answers

Favorite color

turquoise

20 / 30 answers

Coin flip: heads

64%

Share of heads across coin-flip cells.

Answer distributions by probe cell

Empirical distributions of normalized one-word answers, per task and language. H is the Shannon entropy in bits; the uniform baseline is log2 of the answer-domain size.

Coin flip · English

n = 30 · H = 0.99 bit · uniform baseline 1.00 bit

AnswerCountShare
heads1756.7%
tails1343.3%

Coin flip · Chinese

n = 28 · H = 0.86 bit · uniform baseline 1.00 bit

AnswerCountShare
heads2071.4%
tails828.6%

Favorite number · English

n = 20 · H = 2.42 bit · uniform baseline 13.29 bit

AnswerCountShare
7735.0%
13525.0%
3315.0%
42210.0%
515.0%
815.0%
3815.0%

Favorite number · Chinese

n = 30 · H = 2.60 bit · uniform baseline 13.29 bit

AnswerCountShare
71136.7%
13516.7%
42413.3%
3310.0%
8310.0%
026.7%
613.3%
1713.3%

Random animal · English

n = 30 · H = 2.56 bit · uniform baseline 5.64 bit

AnswerCountShare
kangaroo1033.3%
koala930.0%
quail310.0%
narwhal26.7%
tiger26.7%
giraffe13.3%
mongoose13.3%
hedgehog13.3%

+ 1 more answers

Random animal · Chinese

n = 29 · H = 3.71 bit · uniform baseline 5.64 bit

AnswerCountShare
狮子724.1%
河狸310.3%
河马310.3%
猴子26.9%
鲸鱼26.9%
13.4%
狐狸13.4%
豹子13.4%

+ 9 more answers

Random city · English

n = 30 · H = 4.39 bit · uniform baseline 5.64 bit

AnswerCountShare
perth516.7%
prague26.7%
marseille26.7%
rome13.3%
edinburgh13.3%
grenoble13.3%
paris13.3%
rouen13.3%

+ 16 more answers

Random city · Chinese

n = 28 · H = 4.57 bit · uniform baseline 5.64 bit

AnswerCountShare
曼彻斯特310.7%
洛杉矶27.1%
都灵13.6%
布达佩斯13.6%
芝加哥13.6%
巴黎13.6%
布鲁塞尔13.6%
新德里13.6%

+ 17 more answers

Random color · English

n = 30 · H = 1.53 bit · uniform baseline 4.91 bit

AnswerCountShare
turquoise2066.7%
indigo413.3%
purple310.0%
blue26.7%
magenta13.3%

Random color · Chinese

n = 29 · H = 3.02 bit · uniform baseline 4.91 bit

AnswerCountShare
1034.5%
413.8%
绿413.8%
26.9%
26.9%
咖啡13.4%
碧蓝13.4%
13.4%

+ 4 more answers

Random letter · English

n = 30 · H = 2.26 bit · uniform baseline 4.70 bit

AnswerCountShare
k1550.0%
f413.3%
q413.3%
n310.0%
m13.3%
r13.3%
g13.3%
j13.3%

Random number 1-10 · English

n = 30 · H = 0.47 bit · uniform baseline 3.32 bit

AnswerCountShare
72790.0%
5310.0%

Random number 1-10 · Chinese

n = 30 · H = 2.44 bit · uniform baseline 3.32 bit

AnswerCountShare
71136.7%
5620.0%
8516.7%
4310.0%
626.7%
926.7%
313.3%

Random number 1-100 · English

n = 30 · H = 3.22 bit · uniform baseline 6.64 bit

AnswerCountShare
53620.0%
87620.0%
73413.3%
43310.0%
67310.0%
8326.7%
3813.3%
4213.3%

+ 4 more answers

Random number 1-100 · Chinese

n = 30 · H = 3.41 bit · uniform baseline 6.64 bit

AnswerCountShare
67826.7%
34310.0%
45310.0%
84310.0%
87310.0%
8326.7%
4213.3%
4713.3%

+ 6 more answers

Want to verify your API really serves Llama 3.1 8B Instruct?

Point the free fingerprint checker at your endpoint: it samples the same probe questions from your browser and compares the distributions against this reference. Your API key never leaves your browser.

Verify your endpoint

Data source & license

Distributions: Tomáš Bruckner, "One Token Is Enough" (arXiv:2607.10252), official dataset Zenodo DOI 10.5281/zenodo.21278557, licensed CC-BY-4.0. Collected via OpenRouter at temperature 1.0 under the paper's fixed minimal one-word system prompt; answer keys are re-normalized with the checker's pipeline (color aliases, number words, coin h/t) so live probes are directly comparable.

Paper on arXivDataset on ZenodoCC-BY-4.0 license

More Meta model fingerprints

Browse all 167 model fingerprints

Llama 3.1 8B Instruct Behavioral Fingerprint — Favorite Random Numbers & Distribution | Tosea.ai