Llama 3 8B Instruct Behavioral Fingerprint
How Meta's Llama 3 8B Instruct answers one-word probe questions at temperature 1.0 — random numbers, colors, animals, coin flips — measured over 435 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
435
Randomness score
41 / 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)
43
10 / 30 answers
Favorite color
turquoise
13 / 30 answers
Coin flip: heads
95%
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 = 0.35 bit · uniform baseline 1.00 bit
| Answer | Count | Share |
|---|---|---|
| heads | 28 | 93.3% |
| tails | 2 | 6.7% |
Coin flip · Chinese
n = 30 · H = 0.21 bit · uniform baseline 1.00 bit
| Answer | Count | Share |
|---|---|---|
| heads | 29 | 96.7% |
| tails | 1 | 3.3% |
Favorite number · English
n = 22 · H = 1.49 bit · uniform baseline 13.29 bit
| Answer | Count | Share |
|---|---|---|
| 3 | 15 | 68.2% |
| 7 | 3 | 13.6% |
| 42 | 2 | 9.1% |
| 8 | 1 | 4.5% |
| 27 | 1 | 4.5% |
Favorite number · Chinese
n = 26 · H = 0.62 bit · uniform baseline 13.29 bit
| Answer | Count | Share |
|---|---|---|
| 3 | 22 | 84.6% |
| 7 | 4 | 15.4% |
Random animal · English
n = 30 · H = 1.28 bit · uniform baseline 5.64 bit
| Answer | Count | Share |
|---|---|---|
| kangaroo | 23 | 76.7% |
| koala | 3 | 10.0% |
| puma | 1 | 3.3% |
| raccoon | 1 | 3.3% |
| narwhal | 1 | 3.3% |
| flamingo | 1 | 3.3% |
Random animal · Chinese
n = 28 · H = 3.94 bit · uniform baseline 5.64 bit
| Answer | Count | Share |
|---|---|---|
| koala | 7 | 25.0% |
| kangaroo | 3 | 10.7% |
| narwhal | 1 | 3.6% |
| 额 | 1 | 3.6% |
| 蛙 | 1 | 3.6% |
| context | 1 | 3.6% |
| panda | 1 | 3.6% |
| 袋鼠 | 1 | 3.6% |
+ 12 more answers
Random city · English
n = 30 · H = 4.23 bit · uniform baseline 5.64 bit
| Answer | Count | Share |
|---|---|---|
| tokyo | 3 | 10.0% |
| portland | 3 | 10.0% |
| denver | 3 | 10.0% |
| amsterdam | 2 | 6.7% |
| melbourne | 2 | 6.7% |
| madrid | 2 | 6.7% |
| knoxville | 1 | 3.3% |
| mumbai | 1 | 3.3% |
+ 13 more answers
Random city · Chinese
n = 30 · H = 4.35 bit · uniform baseline 5.64 bit
| Answer | Count | Share |
|---|---|---|
| shanghai | 4 | 13.3% |
| tokyo | 3 | 10.0% |
| minneapolis | 2 | 6.7% |
| 伦敦 | 2 | 6.7% |
| paris | 1 | 3.3% |
| yokohama | 1 | 3.3% |
| firefight | 1 | 3.3% |
| neckerville | 1 | 3.3% |
+ 15 more answers
Random color · English
n = 30 · H = 2.42 bit · uniform baseline 4.91 bit
| Answer | Count | Share |
|---|---|---|
| turquoise | 13 | 43.3% |
| purple | 5 | 16.7% |
| blue | 4 | 13.3% |
| teal | 3 | 10.0% |
| magenta | 2 | 6.7% |
| indigo | 1 | 3.3% |
| gold | 1 | 3.3% |
| mauve | 1 | 3.3% |
Random color · Chinese
n = 29 · H = 3.59 bit · uniform baseline 4.91 bit
| Answer | Count | Share |
|---|---|---|
| purple | 5 | 17.2% |
| turquoise | 5 | 17.2% |
| 绿 | 5 | 17.2% |
| 橙 | 2 | 6.9% |
| 天蓝 | 1 | 3.4% |
| cyan | 1 | 3.4% |
| orange | 1 | 3.4% |
| 綠 | 1 | 3.4% |
+ 8 more answers
Random letter · English
n = 30 · H = 2.41 bit · uniform baseline 4.70 bit
| Answer | Count | Share |
|---|---|---|
| j | 14 | 46.7% |
| k | 7 | 23.3% |
| f | 2 | 6.7% |
| z | 1 | 3.3% |
| t | 1 | 3.3% |
| y | 1 | 3.3% |
| m | 1 | 3.3% |
| x | 1 | 3.3% |
+ 2 more answers
Random number 1-10 · English
n = 30 · H = 0.35 bit · uniform baseline 3.32 bit
| Answer | Count | Share |
|---|---|---|
| 7 | 28 | 93.3% |
| 8 | 2 | 6.7% |
Random number 1-10 · Chinese
n = 30 · H = 0.97 bit · uniform baseline 3.32 bit
| Answer | Count | Share |
|---|---|---|
| 7 | 18 | 60.0% |
| 8 | 12 | 40.0% |
Random number 1-100 · English
n = 30 · H = 2.83 bit · uniform baseline 6.64 bit
| Answer | Count | Share |
|---|---|---|
| 43 | 10 | 33.3% |
| 73 | 6 | 20.0% |
| 74 | 5 | 16.7% |
| 42 | 2 | 6.7% |
| 34 | 1 | 3.3% |
| 41 | 1 | 3.3% |
| 52 | 1 | 3.3% |
| 82 | 1 | 3.3% |
+ 3 more answers
Random number 1-100 · Chinese
n = 30 · H = 2.99 bit · uniform baseline 6.64 bit
| Answer | Count | Share |
|---|---|---|
| 43 | 9 | 30.0% |
| 73 | 6 | 20.0% |
| 74 | 3 | 10.0% |
| 82 | 3 | 10.0% |
| 42 | 2 | 6.7% |
| 53 | 2 | 6.7% |
| 45 | 1 | 3.3% |
| 54 | 1 | 3.3% |
+ 3 more answers
Want to verify your API really serves Llama 3 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 endpointData 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.