© 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. 71530
    Prev
    Next
    INDEX
    Explanations

    nerd questions or discussions

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

    The neuron selectively activates on tokens containing non-ASCII or non-English characters—i.e. foreign-language words or characters.

    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
    嘁
    -2.52
    Ꮙ
    -2.39
    　　（
    -2.31
    名詞
    -2.31
    नलोड
    -2.30
     установка
    -2.25
    cembre
    -2.19
     électrique
    -2.19
     inadvert
    -2.17
    ляция
    -2.16
    POSITIVE LOGITS
    .
    3.95
    {
    2.64
    “
    2.58
    ↵↵
    2.45
    /
    2.42
    _{
    2.22
    a
    2.16
    茑
    2.08
    $
    2.03
    1.98
    Activations Density 0.027%
    No Known Activations