© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 102197
    Prev
    Next
    INDEX
    Explanations

    animated show characters

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly activates on capitalized names and titles—especially TV show and character names (e.g. “Bob’s Burgers,” “Homer Simpson,” “Mulder”)—i.e. proper‐noun entities in entertainment contexts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    苼
    -1.54
    </sup>
    -1.52
     داشتم
    -1.43
     that
    -1.38
     those
    -1.37
    . 
    -1.37
    ​.
    -1.33
     there
    -1.27
     گرف
    -1.25
    êmio
    -1.24
    POSITIVE LOGITS
    ways
    1.35
    дости
    1.30
    КС
    1.28
    といった
    1.27
    全家
    1.27
    став
    1.23
     avons
    1.22
    ES
    1.22
    贄
    1.21
     Family
    1.20
    Activations Density 0.008%

    No Known Activations