© 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. 128915
    Prev
    Next
    INDEX
    Explanations

    Yardi, Aldrich, Pollard

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

    The neuron is highly sensitive to mentions of “pericardi-” (e.g. “pericardial” and related heart‐membrane terms).

    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
    ᆞ
    -2.73
    ৢ
    -2.58
    媖
    -2.52
    㩒
    -2.48
    瞜
    -2.48
    苙
    -2.44
    嫑
    -2.38
    蘩
    -2.38
    囖
    -2.38
    
    -2.34
    POSITIVE LOGITS
    as
    2.95
    ?
    2.50
     wanneer
    2.30
    [
    2.30
    what
    2.20
    B
    2.09
    呓
    2.08
     tries
    2.06
    U
    2.06
    G
    2.05
    Activations Density 0.008%

    No Known Activations