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. 16201
    Prev
    Next
    INDEX
    Explanations

    events

    np_max-act · gemini-2.0-flash

    This neuron detects prominent capitalized words that are part of named institutions or event titles (e.g. ‘Rock’, ‘Roll’, ‘Sports’, ‘Chemistry’).

    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
    tet
    -0.07
    iew
    -0.07
     destructive
    -0.07
    attacks
    -0.06
    elters
    -0.06
    ку
    -0.06
    amaha
    -0.06
    aised
    -0.06
     hastalık
    -0.06
    ellas
    -0.06
    POSITIVE LOGITS
     Css
    0.07
    状况
    0.07
     "-",
    0.07
    ={"/
    0.06
     '-',
    0.06
     неб
    0.06
     },{
    0.06
     forall
    0.06
    ügen
    0.06
     Crosby
    0.06
    Activations Density 0.063%

    No Known Activations