© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 5854
    Prev
    Next
    INDEX
    Explanations

    Vatican

    np_max-act · gemini-2.0-flash

    This neuron activates on mentions of the proper‐name “Vatican City,” i.e. the tokens “Vatican” and “City.”

    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
    -service
    -0.06
     Harding
    -0.06
     Luis
    -0.06
     ind
    -0.06
     SHR
    -0.06
    idges
    -0.06
     кле
    -0.05
    _gene
    -0.05
    shade
    -0.05
     disadvantage
    -0.05
    POSITIVE LOGITS
    atican
    0.09
     Vatican
    0.09
     Pope
    0.08
     underwent
    0.07
     pope
    0.07
    azzo
    0.07
     Coronavirus
    0.07
    seau
    0.07
    @register
    0.07
    ван
    0.07
    Activations Density 0.003%

    No Known Activations