© 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. 32877
    Prev
    Next
    INDEX
    Explanations

    conference victory or vote

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

    The neuron fires on multi‐digit numeric tokens—things like game scores, vote counts, percentages, and other statistical numbers.

    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
     sélectionnés
    -1.20
     concernés
    -1.13
     imó
    -1.09
    I
    -1.09
     
    -1.07
     compe
    -1.03
    (
    -1.02
     coste
    -1.02
     hetzelfde
    -1.00
     erm
    -0.99
    POSITIVE LOGITS
     otot
    1.37
     menggel
    1.30
     menatap
    1.28
     éter
    1.28
     meneg
    1.23
    VYMaps
    1.20
     memili
    1.17
    ebly
    1.16
    joda
    1.15
     PartialEq
    1.15
    Activations Density 0.011%

    No Known Activations