© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128326
    Prev
    Next
    INDEX
    Explanations

    The main thing this neuron does is detect numeric data such as statistics, measurements, percentages, and other numbers in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    recent studies and findings

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    😈
    -1.50
    🤠
    -1.38
    😨
    -1.38
    🤪
    -1.29
    👋
    -1.26
    🥶
    -1.25
    龕
    -1.23
    🤫
    -1.23
    🦋
    -1.23
    🔥🔥
    -1.23
    POSITIVE LOGITS
     studies
    1.92
     study
    1.88
     psicología
    1.38
    咻
    1.34
    +/**
    1.31
     Gnade
    1.29
    öffentlichung
    1.29
     researchers
    1.28
     específicamente
    1.28
     ܕ
    1.27
    Activations Density 0.054%

    No Known Activations