© 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. 38413
    Prev
    Next
    INDEX
    Explanations

    numbers or 255

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

    The neuron is specialized to detect the numeric token “255” (the maximum 8-bit value).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    ౭
    -1.80
     requiere
    -1.69
     intric
    -1.66
    䜣
    -1.66
     ."
    -1.64
     trám
    -1.63
    ౩
    -1.58
     Dinas
    -1.58
    -;
    -1.57
     儿童
    -1.55
    POSITIVE LOGITS
     there
    2.55
     if
    1.91
     also
    1.81
     at
    1.80
     Some
    1.69
     while
    1.66
    をはじめ
    1.66
     it
    1.43
     caoutchouc
    1.43
     first
    1.41
    Activations Density 0.024%

    No Known Activations