© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 58374
    Prev
    Next
    INDEX
    Explanations

    dates and weekdays

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

    The neuron is keyed to tokens that are either numbers or all-caps/initial-capital tokens (e.g. numerals and uppercase words).

    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
     in
    -2.08
     then
    -1.91
     this
    -1.76
     on
    -1.70
     it
    -1.68
     all
    -1.64
     what
    -1.63
     because
    -1.60
     just
    -1.60
     so
    -1.59
    POSITIVE LOGITS
     Atención
    1.55
    𨭎
    1.47
     okolí
    1.47
     Momente
    1.45
     intrigu
    1.44
     balení
    1.39
    räder
    1.38
    ᵴ
    1.34
    bedien
    1.33
     คา
    1.33
    Activations Density 0.480%

    No Known Activations