Datasetpublic

lfqian/annulus-eval

Annulus eval sweep results Per-ckpt 3-task eval (PPL pre-2000 year-swap / ECON full 1980–2026 recall+prediction / CAP ICL) for annulus runs v7, v8, v11 … Produced on misha (Yale) via econ_work/ pipeline. results/sweep/ppl_to2005_*.txt — PPL per year-bucket (plain/correct/wnear/wfar) results/econ_ab/ab_*_{recall,prediction}.{json,csv} — P(E) + A–E logits per tau (1980–2026) results/captest_*_{plain,task1}.csv — cap ICL per-task (task1=带[Y2000], plain=不带) dashboard.html — 3-tab… See the full description on the dataset page: https://huggingface.co/datasets/lfqian/annulus-eval.

sourceHugging Faceupdated 1mo agoView on Hugging Face

0likes134downloads

Dataset Card

Annulus eval sweep results

Per-ckpt 3-task eval (PPL pre-2000 year-swap / ECON full 1980–2026 recall+prediction / CAP ICL) for annulus runs v7, v8, v11 … Produced on misha (Yale) via econ_work/ pipeline.

—results/sweep/ppl_to2005_*.txt — PPL per year-bucket (plain/correct/wnear/wfar)
—results/econ_ab/ab_*_{recall,prediction}.{json,csv} — P(E) + A–E logits per tau (1980–2026)
—results/captest_*_{plain,task1}.csv — cap ICL per-task (task1=带[Y2000], plain=不带)
—dashboard.html — 3-tab (PPL/ECON/CAP) interactive summary; sweep_data.json — its data