© 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. 65206
    Prev
    Next
    INDEX
    Explanations

    capital cities

    np_max-act · gemini-2.0-flash

    The neuron activates on tokens in “What is the capital of X?”–style queries and thus detects questions asking for a place’s capital 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
     homem
    -0.06
    Birth
    -0.06
    (ad
    -0.06
     Medicine
    -0.06
    "description
    -0.06
    ************************************************************************
    -0.06
    _small
    -0.06
    庫
    -0.06
     Cyrus
    -0.06
     legit
    -0.06
    POSITIVE LOGITS
    (sel
    0.07
     QPointF
    0.07
     Penalty
    0.07
    <Transform
    0.07
    .getPort
    0.07
    _browser
    0.06
     flooding
    0.06
    <dyn
    0.06
     #=>
    0.06
    manda
    0.06
    Activations Density 0.008%

    No Known Activations