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

    Neuronpedia

    APICircuit 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. 85848
    Prev
    Next
    INDEX
    Explanations

    Tentative language

    np_max-act · gemini-2.0-flash

    This neuron responds to the special metadata and control tokens (e.g. header/separator markers) used to structure the conversation rather than to ordinary words in the user’s 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
    ircraft
    -0.07
     notification
    -0.06
    Employee
    -0.06
     $#
    -0.06
     GetAll
    -0.06
    izing
    -0.06
    ising
    -0.06
    mighty
    -0.06
     piracy
    -0.06
     managerial
    -0.06
    POSITIVE LOGITS
     gew
    0.07
     позволяет
    0.07
     mock
    0.06
    (r
    0.06
    รว
    0.06
    Queryable
    0.06
     definitive
    0.06
    *******/↵↵
    0.06
    .Expressions
    0.06
    *******/↵
    0.06
    Activations Density 0.046%

    No Known Activations