© 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. 128905
    Prev
    Next
    INDEX
    Explanations

    Guggenheim, Guthrie, Roku

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

    The neuron strongly activates on tokens beginning with the letter‐pair “gu” (e.g. Guten, Guggenheim, Gutherie, gutta‐percha).

    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
    of
    -2.66
    al
    -2.63
     ブラウス
    -2.39
     reun
    -2.33
    i
    -2.27
    er
    -2.19
    栳
    -2.19
    ❲
    -2.16
    癭
    -2.14
     بالإضافة
    -2.13
    POSITIVE LOGITS
    ？】
    2.30
    chevêque
    2.22
     tetapi
    2.19
    vají
    2.14
    buzo
    2.11
    を買
    2.11
     altezza
    2.09
    ಞ
    2.08
    codiles
    2.06
    <em>
    2.05
    Activations Density 0.003%

    No Known Activations