© 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. 66444
    Prev
    Next
    INDEX
    Explanations

    separators

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

    This neuron responds to forward‐slash characters (the “/” tokens) used in dates, fractions, citations, or other slash‐separated constructs.

    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
    ８
    -2.88
    Surprisingly
    -2.73
    Only
    -2.64
    Two
    -2.63
    One
    -2.61
    1
    -2.59
    There
    -2.53
     если
    -2.52
    Being
    -2.52
    ２
    -2.50
    POSITIVE LOGITS
    ar
    2.73
    en
    2.47
    m
    2.45
    を始め
    2.39
     และ
    2.39
    in
    2.38
    er
    2.36
    al
    2.34
    änemark
    2.27
     /
    2.25
    Activations Density 0.001%

    No Known Activations