Skip to main content

Claude Haiku 4.5, persona (temperature 0.85, minimal template)

SubPOP · Althing persona · temperature 0.85 · minimal template · 100 questions

synthpanel--claude-haiku-4-5--t0.85--tplminimal--7cffeaef

Metadata

Althing persona · 2 replicates · 100 questions
Method
Althing persona
Base provider
anthropic
Model
claude-haiku-4-5
Dataset
SubPOP
Temperature
0.85
Reasoning effort
default
Template
minimal
Samples / question
30
Questions
100
First replicate
2026-04-11 14:25 UTC
Last replicate
2026-04-11 14:33 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
-74
± 2.2
SPS
SPS, 0 to 1
0.581
± 0.005
Distribution gap
mean JSD, lower is better
0.406
± 0.021
Rank agreement
Kendall τ, −1 to 1
0.226
± 0.003

Variance summary

Replicates
2
Std dev
2.2 Index points (SPS 0.005)
Range
Index -76 to -73 (SPS 0.577–0.584)
Coefficient of variation (SPS)
0.0090

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 (2)

Run Timestamp Index SPS JSD Kendall τ N
subpop_synthpanel_openrouter_anthropic_claude-haiku-4-5 t=0.85 tpl=minimal_20260411_142531 2026-04-11 14:25 UTC -73 0.584 0.421 0.228 100
subpop_synthpanel_openrouter_anthropic_claude-haiku-4-5 t=0.85 tpl=minimal_20260411_143319 2026-04-11 14:33 UTC -76 0.577 0.391 0.224 100