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

    Neuronpedia

    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. 44595
    Prev
    Next
    INDEX
    Explanations

    the

    np_max-act · gemini-2.0-flash

    This neuron detects the user’s instruction prompts that begin with “Given the document below” (i.e. the meta‐task framing lines).

    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
    ıştır
    -0.07
    _hd
    -0.07
    ¡
    -0.07
    �
    -0.07
     Osaka
    -0.06
     login
    -0.06
    atus
    -0.06
     identities
    -0.06
     ❤
    -0.06
     engaged
    -0.06
    POSITIVE LOGITS
     yeter
    0.07
    .Print
    0.06
     viên
    0.06
    dress
    0.06
     весь
    0.06
     DOWNLOAD
    0.06
    rubu
    0.06
     点击
    0.06
     refs
    0.06
     cosa
    0.06
    Activations Density 0.002%

    No Known Activations