claim
active
claim:the-scaling-law-in-jones-et-al-may-reflect-prompt-harmfulness-distribution-more-than-the-model-s-output-distributionThe scaling law in Jones et al. may reflect prompt-harmfulness distribution more than the model's output distribution
Distinguishes their setting from this one.
Neighborhood — ranked by edge-count
Papers (1)
paper
- Predicting Rare LLM Failures with 30× Fewer Rolloutscitesmentions
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.
- Cited hypothesis from Lin et al. 2022 suggesting larger models become more capable of deception
- Core phenomenological argument.
- Compute-optimal hyperparameters follow predictable power-law relationships.
- Empirically observed power law relationship between data scale and model performance; supports convergence hypothesis
- Scaling Laws for Activation Steering with Llama 2 Models and Refusal Mechanisms (Ali et al., 2025)concept0.766Related work finding larger models more resistant to steering, potentially consistent with ESR in 70B
- Observation that SAE loss decreases as a power law with compute budget.
- Acknowledges the confound of not explicitly instructing models to track wealth, yet points to reasoning gaps given code agents avoid errors without prompts.
- Hypothesis cited in paper suggesting deceptive capabilities may scale with model size