© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 19978
    Prev
    Next
    INDEX
    Explanations

    ADVERTISEMENT or xxx

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

    This neuron fires on standalone label‐like tokens—e.g. all‐caps words (“ADVERTISEMENT,” “DDT”), CamelCase identifiers (“SpringBootTest”), and similar single‐token headings or code 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
    是在
    -2.50
    全都
    -2.36
    in
    -2.31
     obstruct
    -2.31
     unsuspecting
    -2.22
     glaring
    -2.20
     Jezus
    -2.17
     están
    -2.16
    也可
    -2.14
    or
    -2.13
    POSITIVE LOGITS
    酲
    2.52
    ྅
    2.42
    対応
    2.42
    𓆜
    2.34
    蹌
    2.33
     mereka
    2.28
    盌
    2.27
    媜
    2.25
    🤠
    2.23
    牒
    2.22
    Activations Density 0.002%

    No Known Activations