© 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. 21071
    Prev
    Next
    INDEX
    Explanations

    markdown

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

    The neuron is picking out named entities (proper nouns), e.g. place names like “Kaduna” or technology/package names like “markdown.”

    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
    '
    -4.03
    ing
    -2.97
    *
    -2.66
    0
    -2.64
    =
    -2.61
    8
    -2.45
     errone
    -2.38
    M
    -2.34
    ’
    -2.27
    At
    -2.25
    POSITIVE LOGITS
    lossians
    2.75
    इटम
    2.70
     Änderungen
    2.61
    璈
    2.61
     llevaron
    2.42
    硖
    2.38
     natuurlijke
    2.33
     zwarte
    2.31
    驽
    2.31
     unieke
    2.30
    Activations Density 0.002%

    No Known Activations