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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 58978
    Prev
    Next
    INDEX
    Explanations

    Napoleon

    np_max-act · gemini-2.0-flash

    The neuron primarily activates on subword tokens composing “Napoleon” (as in “Napoleon Bonaparte”), effectively detecting mentions of Napoleon.

    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
     critic
    -0.07
     lb
    -0.06
    Conditions
    -0.06
     Detroit
    -0.06
     رسمی
    -0.06
     executives
    -0.05
     pulp
    -0.05
     tween
    -0.05
     InvalidArgumentException
    -0.05
     textView
    -0.05
    POSITIVE LOGITS
     Napoleon
    0.10
    oleon
    0.08
    .timezone
    0.07
     verdad
    0.07
    ucht
    0.07
    άβ
    0.07
    ической
    0.06
    態
    0.06
    atos
    0.06
     sail
    0.06
    Activations Density 0.006%

    No Known Activations