Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 84375
    Prev
    Next
    INDEX
    Explanations

    born

    np_max-act · gemini-2.0-flash

    The neuron fires on numeric tokens that represent dates—especially birth years, days, and months—highlighting date‐related numbers in the text.

    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
    BW
    -0.06
    潮
    -0.06
     مون
    -0.06
     ngủ
    -0.06
    .black
    -0.06
    QUAL
    -0.06
     Vote
    -0.06
    Đây
    -0.06
     підс
    -0.06
    .css
    -0.06
    POSITIVE LOGITS
     geom
    0.07
    (logging
    0.07
     Lets
    0.06
    Tiles
    0.06
    ади
    0.06
    .providers
    0.06
     Writers
    0.06
     Providers
    0.06
    gunta
    0.06
     usually
    0.06
    Activations Density 0.013%

    No Known Activations