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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 38230
    Prev
    Next
    INDEX
    Explanations

    used

    np_max-act · gemini-2.0-flash

    The neuron fires on technical or domain‐specific tokens—especially acronyms, product or model names, and similarly “jargon” words (e.g. IO-Geräte, Somfy, file, C++, etc.).

    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
    .dt
    -0.07
    ět
    -0.07
     které
    -0.07
     IBM
    -0.06
    ์ช
    -0.06
     není
    -0.06
    Deleted
    -0.06
    Director
    -0.06
    UserName
    -0.06
    /datatables
    -0.06
    POSITIVE LOGITS
     sanitize
    0.06
     mse
    0.06
    _clk
    0.06
    ursion
    0.06
    itlement
    0.06
     extingu
    0.06
     tapered
    0.06
     Edgar
    0.06
    .thread
    0.06
     pca
    0.06
    Activations Density 0.088%

    No Known Activations