method
active
method:log-linear-extrapolation-stepLog-linear extrapolation step
Fits log compliance vs α on α with compliance <50% and extrapolates to α=1.
Neighborhood — ranked by edge-count
Papers (1)
paper
Methods (2)
method
- Logit interpolation steprelated_toSamples from interpolated logits of original and abliterated models at α.
- The paper's method: interpolate between two models in logit space, measure compliance where common, and extrapolate to the original.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Parameter-free loss transformation applied to each task loss to equalize scales
- Stated limitation.
- Dot product of summed decoder vector with unembedding vector, following Hanna et al.
- Mathematical equivalence showing logarithm transformation recovers IMTL-L in the limit
- Correlative technique measuring the type of information encoded in distributed representations via linear predictability.
- The incremental unfolding characteristic of morphogenesis, where each step arises from the previous state.
- log P(M(α) complies) is approximately linear in α when probability is not close to 1 on Qwen 3 4B pathfinding0.731Empirical basis for the extrapolation.
- Headline result (Figure 1).