finding
active
finding:best-clt-on-gpt-2-small-achieves-l0-of-79-and-59-token-accuracy-consistent-with-ameisen-et-alBest CLT on GPT-2 Small achieves L0 of 79 and 59% token accuracy, consistent with Ameisen et al.
Training quality of the CLT.
Source paper
extracted_fromNeighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Main conclusion.
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.
- Training configuration.
- Evidence humans and AI use different representations.
- GPT-4o-mini most consistent in reproducing persona-aligned atomic distributions across repeated generations
- Early large language model cited as an example of transformer-based LLMs
- GPT-4.1 robustness collapse values
- Specification of AI models used in the two pilot experiments
- Best performing VS variant for math synthetic data generation with GPT-4.1
- GPT-3.5-turbo produces most internally coherent persona-aligned generations