© 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. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 7580
    Prev
    Next
    INDEX
    Explanations

    emojis and specific terms

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

    The neuron is primarily detecting the presence of emoji and other non-alphabetic/pictographic Unicode symbols.

    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
     an
    -2.33
     with
    -2.33
     For
    -2.31
     to
    -2.27
     large
    -2.09
     on
    -2.09
     blatantly
    -2.09
     there
    -2.08
     after
    -2.03
     Get
    -2.03
    POSITIVE LOGITS
     harten
    2.22
     quintessential
    2.14
    selbe
    2.03
    圌
    2.03
     MICHAEL
    2.03
    懵
    2.02
    蹰
    2.00
     ネック
    1.98
     fyra
    1.93
     ホルダー
    1.91
    Activations Density 0.005%

    No Known Activations