Abliteration at 3-bit: no disposition effect, a model-dependent verbosity shift

Final study. Tests whether the disposition effects reported for abliterated models at full precision survive low-bit quantization and long input. Two model sizes (27B and 125B), two abliterated builds, four context lengths from 4k to 200k tokens, 5712 generations. No detectable difference in optimism, stated confidence or hedging at either size; no drift with context length; a verbosity effect whose direction reverses between model sizes (Cliff delta -0.33 and -0.44 at 27B); a chat-template audit returning exactly zero for bare user turns with the override mechanism verified; speculative decoding shown to preserve outputs exactly. Includes probe generator, runner, analysis code, raw generations, template audit, prior-art sweeps, the preregistration and its amendment.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-12
DOI
https://doi.org/10.5281/zenodo.24166077
Primary Topic
Natural Language Processing Techniques
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Abliteration at 3-bit: no disposition effect, a model-dependent verbosity shift

Jackson Peddy
1 links
Zenodo (CERN European Organization for Nuclear Research)
Natural Language Processing Techniques
preprint

Abliteration at 3-bit: no disposition effect, a model-dependent verbosity shift

Jackson Peddy
preprint en

Abstract

Final study. Tests whether the disposition effects reported for abliterated models at full precision survive low-bit quantization and long input. Two model sizes (27B and 125B), two abliterated builds, four context lengths from 4k to 200k tokens, 5712 generations. No detectable difference in optimism, stated confidence or hedging at either size; no drift with context length; a verbosity effect whose direction reverses between model sizes (Cliff delta -0.33 and -0.44 at 27B); a chat-template audit returning exactly zero for bare user turns with the override mechanism verified; speculative decoding shown to preserve outputs exactly. Includes probe generator, runner, analysis code, raw generations, template audit, prior-art sweeps, the preregistration and its amendment.

Zenodo (CERN European Organization for Nuclear Research)
Natural Language Processing Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon