© 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. 129610
    Prev
    Next
    INDEX
    Explanations

    {

    np_max-act · gemini-2.0-flash

    The neuron responds to C/C++ class‐declaration boilerplate (e.g. the “class X : public Y” line and related macros like Q_OBJECT).

    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
     reinterpret
    -0.07
    &R
    -0.07
    .prefix
    -0.07
    [C
    -0.06
     شده
    -0.06
    벤
    -0.06
    licht
    -0.06
     camar
    -0.06
     ZX
    -0.06
    '),↵
    -0.06
    POSITIVE LOGITS
    (notification
    0.08
    -fontawesome
    0.08
     Economy
    0.07
     prejudices
    0.07
    .struts
    0.06
    uuml
    0.06
     Newsp
    0.06
     Valve
    0.06
     luận
    0.06
     GUIDATA
    0.06
    Activations Density 0.004%

    No Known Activations