© 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. 110771
    Prev
    Next
    INDEX
    Explanations

    browsers

    np_max-act · gemini-2.0-flash

    The neuron is tuned to detect mentions of web browsers—especially browser names and compatibility warnings (e.g. “unsupported browser,” “Internet Explorer,” “Firefox,” “Chrome”).

    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
    038
    -0.07
     کمتر
    -0.07
     зада
    -0.06
    ColumnsMode
    -0.06
    blings
    -0.06
     Tao
    -0.06
    каж
    -0.06
     Temmuz
    -0.06
     değiştir
    -0.06
    YPD
    -0.06
    POSITIVE LOGITS
    ifax
    0.07
     ranking
    0.07
     project
    0.07
    pagen
    0.07
    _Widget
    0.06
    MatrixMode
    0.06
    -ever
    0.06
     Qt
    0.06
    ある
    0.06
     SITE
    0.06
    Activations Density 0.002%

    No Known Activations