Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 42428
    Prev
    Next
    INDEX
    Explanations

    Code/Formulas

    np_max-act · gemini-2.0-flash

    This neuron detects occurrences of the word “causes” in the causal‐narrative output format.

    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
    _every
    -0.07
     with
    -0.07
     coach
    -0.07
     Agents
    -0.07
     velocities
    -0.07
     Ker
    -0.07
     portals
    -0.06
     deze
    -0.06
    .reporting
    -0.06
    -print
    -0.06
    POSITIVE LOGITS
    ิ้
    0.07
     застосування
    0.07
    تبة
    0.07
    PositiveButton
    0.06
     nhấn
    0.06
    ancock
    0.06
     فرآ
    0.06
     встанов
    0.06
    .bluetooth
    0.06
    ogonal
    0.06
    Activations Density 0.022%

    No Known Activations