© 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. 10567
    Prev
    Next
    INDEX
    Explanations

    stay followed by prepositions or actions

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

    The neuron responds specifically to the token “stay” (in any case), marking every occurrence of that exact word.

    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
    鑁
    -2.81
    uſe
    -2.81
    靺
    -2.80
    郿
    -2.80
    峩
    -2.78
    玊
    -2.70
    鈊
    -2.67
    ൎ
    -2.52
     tornar
    -2.52
    薅
    -2.50
    POSITIVE LOGITS
    i
    3.27
    {
    3.23
    ra
    2.73
    The
    2.56
    *
    2.53
    ating
    2.52
    in
    2.47
    el
    2.47
    at
    2.42
    out
    2.38
    Activations Density 0.008%

    No Known Activations