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. 84672
    Prev
    Next
    INDEX
    Explanations

    Config

    np_max-act · gemini-2.0-flash

    The neuron responds to numeric literal tokens—especially decimal numbers—within XML attributes.

    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
     homeschool
    -0.07
    _launcher
    -0.07
    �
    -0.06
     urn
    -0.06
    -icons
    -0.06
     Pasta
    -0.06
    Domain
    -0.06
     Nan
    -0.06
    $text
    -0.06
    List
    -0.06
    POSITIVE LOGITS
     incess
    0.07
     els
    0.06
     fours
    0.06
     harder
    0.06
    дия
    0.06
    cessive
    0.06
     ViewData
    0.05
     væ
    0.05
     člověk
    0.05
    izontally
    0.05
    Activations Density 0.007%

    No Known Activations