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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 12918
    Prev
    Next
    INDEX
    Explanations

    do

    np_max-act · gemini-2.0-flash

    The neuron responds to mentions of divine attributes—especially “omni-” terms like omnipotence or omniscience.

    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
    留
    -0.07
    plevel
    -0.06
    etag
    -0.06
     captivity
    -0.06
    برای
    -0.06
     subplot
    -0.06
     Unicorn
    -0.06
     complicated
    -0.06
     metre
    -0.06
    chang
    -0.06
    POSITIVE LOGITS
     Otto
    0.06
     ساعت
    0.06
    務
    0.06
    乾
    0.06
     bảo
    0.06
    180
    0.06
    /terms
    0.06
     işi
    0.06
    iang
    0.06
     newY
    0.06
    Activations Density 0.029%

    No Known Activations