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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 85027
    Prev
    Next
    INDEX
    Explanations

    school/team names

    np_max-act · gemini-2.0-flash

    This neuron detects capitalized sports team names (often color-plus-mascot combinations) 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
    --------↵↵
    -0.07
     (>
    -0.06
     vi
    -0.06
    _ud
    -0.06
    _collision
    -0.06
     Metodo
    -0.06
    opencv
    -0.06
     //////////////////////////////////////////////////////////////////////
    -0.06
    ANCELED
    -0.06
    _lib
    -0.06
    POSITIVE LOGITS
     Wildcats
    0.06
    hawks
    0.06
     }</
    0.06
     Т
    0.06
     lectures
    0.06
    [src
    0.06
    AMPL
    0.06
     béné
    0.06
    되어
    0.06
     awaiting
    0.06
    Activations Density 0.010%

    No Known Activations