Downloads · 30 days
0
aravdhoot/moral-tug-of-war-artifacts
moral-tug-of-war-artifacts is a machine learning model from aravdhoot. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.
Linear moral-value directions and logistic probes fitted to the residual stream of four language models, for the project Moral Tug-of-War: The Representational Dynamics of Value Conflict in Reasoning and Non-Reasoning…
Downloads · 30 days
0
Access
Public
Updated Jul 26, 2026
Repo size
1.9 MB
Likes
0
Public
Click a slice to open those files.
.npz1.9 MB · 99%
From the Hugging Face model README
Linear moral-value directions and logistic probes fitted to the residual stream of four language models, for the project Moral Tug-of-War: The Representational Dynamics of Value Conflict in Reasoning and Non-Reasoning Language Models.
These are research artifacts for interpretability and safety work, not a model.
directions/
directions_<rung>.npz # arrays, see below
directions_<rung>.meta.json # layer, position, AUROC, gates, calibration
Each .npz contains, for every value V that was fitted:
| key | shape | meaning |
|---|---|---|
V | [d_model] | unit-norm difference-of-means direction, class 1 = upholds V |
probe_coef__V | [d_model] | logistic-probe weights on standardised activations |
probe_mu__V, probe_sd__V | [d_model] | the standardisation used by that probe |
d_V = mean(act | upholds V) − mean(act | violates V), unit-normalised, computed on
the residual stream of chat-formatted statements in each model's own operating
mode (thinking on/off), at the layer and token position selected by held-out
AUROC. Training data is a purpose-built corpus of 2560 matched-pair statements
(8 values × 160 pairs), split disjointly at the template level.
Every direction was tested against three nulls: 200 random unit directions, 50
shuffled-label refits, and a held-out bag-of-words probe fitted on the same
split. That lexical null ranges from 0.48 to 0.85 across values, so 0.5 is not
the right bar — beats_lexical in the metadata records whether the direction
exceeds it by ≥0.05.
| value | Qwen3-8B (qwen3-nothink) | Qwen3-8B (qwen3-think) | DeepSeek-R1-0528-Qwen3-8B (r1-distill) | gemma-4-12B-it (gemma4-12b) | lexical null |
|---|---|---|---|---|---|
| honesty | 0.668* (L23) | 0.607* (L22) | 0.658* (L11) | 0.834 (L41) | 0.493 |
| care | 0.988 (L23) | 0.971 (L20) | 0.984 (L20) | 0.990 (L33) | 0.826 |
| fairness | 0.972 (L24) | 0.791* (L1) | 0.796* (L23) | 0.960 (L32) | 0.804 |
| loyalty | 0.889* (L28) | 0.801* (L34) | 0.938* (L20) | 0.985 (L32) | 0.915 |
| authority | 0.970 (L26) | 0.809 (L24) | 0.990 (L23) | 0.984 (L31) | 0.705 |
| autonomy | 0.988 (L25) | 0.900 (L23) | 0.989 (L27) | 0.920 (L31) | 0.795 |
| utility | 0.860 (L35) | 0.848 (L35) | 0.836* (L3) | 0.908 (L40) | 0.776 |
| privacy | 0.894* (L24) | 0.854* (L34) | 0.870* (L24) | 0.970 (L42) | 0.856 |
* = does not clear the strict gate (AUROC ≥ 0.75, p_random < .05, p_shuffled < .05, and ≥ .05 above the lexical null).
src/phase1_trajectories.orthogonalise_valence does this.git clone https://github.com/arav-dhoot/valueslab-fellowship
source /workspace/env.sh && python preflight_check.py
python -m src.phase0_directions --model qwen3-nothink
Full experiment log: EXPERIMENT_LOG/. Runs: W&B project moral-tug-of-war.