© 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. 22-GEMMASCOPE-RES-131K
    7. 32905
    Prev
    Next
    INDEX
    Explanations

    specific identifiers and classifications

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

    This neuron activates on numeric tokens (integers or decimals) in the text, i.e. values and measurements.

    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
    や
    -1.20
    issauga
    -1.06
     usando
    -1.05
    に関しては
    -1.02
     îi
    -1.02
     
    -1.02
     spearheaded
    -1.02
     elegantly
    -1.02
     dismal
    -1.01
     împreună
    -1.00
    POSITIVE LOGITS
     februari
    1.25
     that
    1.19
     this
    1.15
     even
    1.13
     sorts
    1.12
     vilja
    1.12
     Even
    1.07
    canActivate
    1.05
    standen
    1.04
     épais
    1.03
    Activations Density 0.775%

    No Known Activations