© 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. 59742
    Prev
    Next
    INDEX
    Explanations

    crunchy texture and "crunch time"

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

    The neuron responds to occurrences of the substring “crunch” (e.g. “crunch,” “crunching,” or as part of “TechCrunch”).

    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
    쫀
    -3.50
    </h3>
    -2.88
    냑
    -2.77
    盌
    -2.67
    猻
    -2.67
    訁
    -2.63
    癭
    -2.52
     This
    -2.52
     Outros
    -2.48
    킁
    -2.45
    POSITIVE LOGITS
    </b>
    2.89
    <bos>
    2.48
    me
    2.34
    $
    2.28
    F
    2.28
    And
    2.19
     comien
    2.17
    𝐒
    2.14
     ويت
    2.11
    های
    2.09
    Activations Density 0.002%

    No Known Activations