Skip to main content

Llama 3.3 70B, persona

SubPOP · Althing persona · default settings · 100 questions

synthpanel--llama-3.3-70b-instruct--tdefault--tplcurrent--9d6e5c2c

Metadata

Althing persona · 1 replicate · 100 questions
Method
Althing persona
Base provider
meta
Model
llama-3.3-70b-instruct
Dataset
SubPOP
Temperature
default
Reasoning effort
default
Template
current
Samples / question
15
Questions
100
First replicate
2026-07-15 17:17 UTC
Last replicate
2026-07-15 17:17 UTC

Aggregated metrics

Mean across replicates, ± sample standard deviation (shown only with 2+ replicates). SynthBench Index: 0 = random answers, 100 = real people. SPS is the raw 0 to 1 score it is computed from. Index = 100 × (SPS − SPS of random answers) ÷ (human ceiling − SPS of random answers), on this config's dataset. The leaderboard averages it over the core datasets.

Index
0 = random answers, 100 = real people
8
single replicate
SPS
SPS, 0 to 1
0.776
single replicate
Distribution gap
mean JSD, lower is better
0.363
single replicate
Rank agreement
Kendall τ, −1 to 1
0.470
single replicate

Variance summary

Replicates
1
Std dev
—
Range
—
Coefficient of variation (SPS)
—

Per-topic breakdown

Index per topic (mean across replicates; 0 = random answers, 100 = real people). All seven topics are shown; topics this config's questions don't cover are marked "no questions".

Show data table

Loading data…

Replicates (1)

Run Timestamp Index SPS JSD Kendall τ N
subpop_synthpanel_openrouter_meta-llama_llama-3.3-70b-instruct_20260715_171744 2026-07-15 17:17 UTC 8 0.776 0.363 0.470 100