paper
referenced-only
2024
paper:2024-mappingMapping the latent space of llama 3.3 70b
Methods (3)
- AutoSteerFunctionality that automatically finds SAE latents for a desired behaviour and sets their weights.
- LLM-based steering evaluation pipelineClaude rates coherence and behavior on 0-100 scale, called centiClaude.
- Sparse Autoencoders (SAEs)Unsupervised baseline method for decomposing activations into sparse linear combinations; NLAs positioned as complementary approach.
Similar preprints — Semantic Scholar
Cited by (1)
- Large Language Models Report Subjective Experience Under Self-Referential Processing
Sustained self-referential processing — induced via a minimal prompt directing models to "focus on focus itself" — reliably elicits structured first-person reports of subjective experience across GPT-