Llama 3.1 405B

Large open-weight model showing compliance gap in helpful-only setting

Neighborhood — ranked by edge-count

concept

LLaMA 3.3 70B
related_to
The model used in Experiment 2 for SAE feature steering experiments via Goodfire API
LLaMA3.1-8B
related_to
One of four LLMs selected for representation analysis; embedding dimension D=4096; used as demonstration model in scatter plots.
LLaMA3.1-70B
related_to
One of four LLMs selected; larger model with D=8192 embedding dimension; analyzed across proportionally aligned layers.

cosine ≥ 0.65 · no typed edge

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

Llama-3.1-8B-Instructconcept0.839
Primary qualitative demonstration model and one of 14 LLMs benchmarked
Llama-3.2-3B-Instructconcept0.831
3B Llama model tested; used for injection stride visualization
Meta-Llama-3.1-8B-Instructconcept0.816
Backbone model used in E3 geometry analysis.
Llama-3.2-1B-Instructconcept0.808
Smallest Llama model tested; benchmarked across all injection methods
Llama-3.3-70B-Instructconcept0.807
Primary model of interest showing substantial ESR; largest model tested in the study
LLaMA / LLaMA2 / LLaMA3concept0.801
Language model family used in cross-modal alignment experiments across multiple sizes
Llama 2concept0.785
Meta's open large language model cited as an example of the class of models under discussion
Llama 3.1 405B shows 14% compliance gap in minimal helpful-only replication but smaller Llama and Mistral models show no gapfinding0.778
Replication across open-weight models supports scale-emergence finding