© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 103777
    Prev
    Next
    INDEX
    Explanations

    one half of

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

    The neuron fires on mentions of a “half of” or fractional‐half construct (e.g. “one half of X”).

    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
     have
    -1.58
     in
    -1.25
     see
    -1.23
    deleteById
    -1.19
    ↵
    -1.18
     to
    -1.17
     adjust
    -1.16
    8
    -1.16
     doméstica
    -1.16
    ”?
    -1.16
    POSITIVE LOGITS
     Когда
    1.24
     a
    1.23
    include
    1.18
    on
    1.18
    裊
    1.17
    ﹤
    1.16
     échange
    1.16
    izacion
    1.16
    Два
    1.16
     icônes
    1.13
    Activations Density 0.113%

    No Known Activations