© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 126282
    Prev
    Next
    INDEX
    Explanations

    negotiations and agreements

    np_max-act · gemini-2.0-flash

    The neuron responds to formal legal or governmental language—titles, institutional roles, and related terms (e.g. “Attorney General,” “settlement,” “helped secure”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     film
    -0.07
     francouz
    -0.06
    }';↵
    -0.06
    960
    -0.06
     Today
    -0.06
     skiing
    -0.06
    Construct
    -0.06
     Sn
    -0.06
    _categorical
    -0.06
     Outcome
    -0.06
    POSITIVE LOGITS
    .script
    0.06
     TORT
    0.06
    Numeric
    0.06
     Smoking
    0.06
     physically
    0.06
     enterprises
    0.06
     ø
    0.06
    .scrollHeight
    0.06
     (-
    0.06
    istream
    0.06
    Activations Density 0.062%

    No Known Activations