© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 71913
    Prev
    Next
    INDEX
    Explanations

    staff, cast, team, band members

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

    This neuron detects numeric tokens—especially standalone numbers such as dates, years, counts, or other quantifiers—in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     can
    -1.54
     while
    -1.52
     or
    -1.52
     in
    -1.38
     more
    -1.34
     if
    -1.31
     through
    -1.30
     after
    -1.26
     provide
    -1.20
     perhaps
    -1.17
    POSITIVE LOGITS
     who
    2.09
     whom
    1.47
     gồm
    1.43
     acte
    1.37
     eléct
    1.34
    whom
    1.32
     chè
    1.31
     anadolu
    1.30
     pittores
    1.29
    cluding
    1.28
    Activations Density 0.061%

    No Known Activations