© 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 EvalsExportsinterp-engineNEWAPI 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. 58024
    Prev
    Next
    INDEX
    Explanations

    article snippets

    np_max-act · gemini-2.0-flash

    text related to job postings and professional announcements.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron is primarily activated by title-style or navigation labels (e.g. single‐word site headings like “Event,” “News,” “Listing,” “Submit,” “Calendar,” 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
     invaded
    -0.07
    =min
    -0.07
    political
    -0.07
     sided
    -0.07
    ذار
    -0.07
     Lenovo
    -0.06
     fj
    -0.06
     oleh
    -0.06
     nord
    -0.06
     algum
    -0.06
    POSITIVE LOGITS
    (AdapterView
    0.08
     CreateUser
    0.07
    .MapPath
    0.07
    .getResponse
    0.06
    ảnh
    0.06
    _QUAL
    0.06
    осудар
    0.06
     혹
    0.06
    เหต
    0.06
     Profes
    0.06
    Activations Density 0.075%

    No Known Activations