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. 130791
    Prev
    Next
    INDEX
    Explanations

    special characters

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on floating‐point numeric literals (decimal numbers with fractional parts).

    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
     Zaman
    -0.07
     dnů
    -0.06
    که
    -0.06
     fertilizer
    -0.06
    .preferences
    -0.06
     jeg
    -0.06
    Simon
    -0.06
    ливі
    -0.06
     precursor
    -0.06
    xffffff
    -0.06
    POSITIVE LOGITS
    :::::|
    0.07
    aho
    0.07
     uy
    0.06
    학과
    0.06
     Observ
    0.06
    -insert
    0.06
    :C
    0.06
     아주
    0.06
    �
    0.06
    ,state
    0.06
    Activations Density 0.048%

    No Known Activations