© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 65626
    Prev
    Next
    INDEX
    Explanations

    whisky and whiskey

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

    The neuron activates on tokens containing the substring “whisk,” e.g. whiskers, whisked, whisky/whiskey.

    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
    ”
    -2.97
    秝
    -2.69
    »,
    -2.55
    He
    -2.47
     positioning
    -2.44
    ()
    -2.42
     anticipo
    -2.34
     exceptionally
    -2.33
    겔
    -2.27
    va
    -2.25
    POSITIVE LOGITS
    مشخصات
    2.52
     That
    2.41
    蕭
    2.28
     که
    2.20
     Это
    2.19
    März
    2.13
    ствием
    2.09
     whisk
    2.09
     Rooms
    2.08
     déficit
    2.06
    Activations Density 0.004%

    No Known Activations