© 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. 84867
    Prev
    Next
    INDEX
    Explanations

    code and medical terms

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

    empty lines or whitespace in technical and scientific documents.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron primarily activates on superscript citation markers (the caret “^” used for footnotes/references).

    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
    </strong>
    -1.89
     灵
    -1.76
    :
    -1.74
    -
    -1.72
    溅
    -1.70
    過ぎる
    -1.69
    ,
    -1.63
     свою
    -1.59
    berfläche
    -1.52
    髒
    -1.52
    POSITIVE LOGITS
    ⃙
    1.65
     including
    1.60
     now
    1.60
     further
    1.48
     eventually
    1.48
    üenza
    1.46
     uveden
    1.43
    本科
    1.42
     następnie
    1.41
    特殊
    1.41
    Activations Density 0.000%

    No Known Activations