© 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. 21569
    Prev
    Next
    INDEX
    Explanations

    Roanoke, manipulators, Karim, MyBatis

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

    The neuron is primarily sensitive to uncommon proper names or specialized technical terms (e.g. place names, personal names, project names, and niche domain words).

    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
     But
    -3.78
    但我
    -3.13
    但
    -2.83
     tetapi
    -2.67
     nhưng
    -2.61
     If
    -2.59
    用
    -2.56
     pero
    -2.56
    '
    -2.55
    但是
    -2.48
    POSITIVE LOGITS
    狆
    3.83
    艫
    3.36
    蘗
    3.25
    ating
    3.23
    푆
    3.00
    鯔
    2.98
    婠
    2.94
    腼
    2.91
    勐
    2.83
    ᠁
    2.72
    Activations Density 0.022%

    No Known Activations