Marnie — Analytics
Production
Analytics
Research
Total Runs
Completed
Total Tokens
Avg Loops
Convergence Rate
Avg Runtime
seconds
Runs per day (30 days)
By Condition
ConditionRunsAvg TokensConv.%
Loading…
By Genre
GenreRuns
Loading…
Loading…

B2 — Inter-Rater Reliability tracking. Upload sampled critique labels (agent vs. human) to compute Cohen's κ.

Batch IDn Critiques Cohen's κAgreement RateDate
Loading…

B3 — Statistical test results. Enter results from Wilcoxon signed-rank or Mann-Whitney U tests. Results auto-format for paper citation.

TestConditionsMetric Statisticp-value Effect (r)95% CIn A / n B Citation
No results yet.
Datasets

Each dataset is a group of runs sharing a source label. Deleting a dataset removes all its runs.

Loading…
Add or update runs

Paste a JSON array of run records (or upload a .json file). Existing run_ids are updated; new ones are inserted. Set a dataset label to stamp every imported row with the same source.

Browse & remove
Run IDDatasetConditionLeadManuscript ACI
Search to list runs.

Download a fully self-contained study report (works offline). Source is locked to Testing for study exports.

Run counts by condition (testing)

Completed runs from your admin account. Select one or more runs, choose a run type, and assign them to the analytics dashboard.

Run Date Loops Assignment Actions
Loading…