Do the findings about MDS injection effectiveness generalize to base (non-instruction-tuned) language models?

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• MDS injections align with the Linear Representation Hypothesis: target trait varies near-linearly with alpha in open-ended generationclaim0.779
  Theoretical alignment claim backed by OLS R2 analysis showing 96.15% of trends have R2>=0.75
• DAS finds causal effect at all training timesteps including when model is just initialisedfinding0.773
  Corroborates Wu et al. 2023 finding that DAS expressivity inflates causal effect estimates
• MDS injections show no salient patterns in MPI-120 inventory responses beyond occasional co-occurring peaksfinding0.765
  Contrasts with SJT results; leads authors to narrow analyses to SJT responses
• We hypothesize that native self-report, fine-tuned introspection models, and trained activation-to-language systems will show different performance on bias-resistant localization and strength benchmarkshypothesis0.764
  Comparative prediction motivating future work contrasting different approaches to LLM self-knowledge
• Simple difference-in-mean probes generalize as well as other probing techniques while identifying directions which are more causally implicated in model outputsclaim0.761
  Key methodological claim: MM probes are both competitive in accuracy and superior in causal influence
• Results may not fully generalize to all models and scenarios because the model organism relies on hints and nudges and Llama Nemotron cannot consistently distinguish evaluation/deployment based on subtle cuesclaim0.758
  Key limitation acknowledged by authors.
• OCEAN MDS injection covariance patterns departing from the Big Two model suggest a gap between learned LLM representations and human psychologyclaim0.750
  Interpretive conclusion from Big Two mismatch finding; tentative due to only 46.15% match rate
• It's tricky, because for a typical language model the entity is sort of tricameral: the base simulator, the simulated simulator, and the simulated awareness.quote0.748
  Antra's earlier definitive statement of the tricameral model.