© 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. 33851
    Prev
    Next
    INDEX
    Explanations

    reported by authority

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

    The neuron fires on numeric expressions and figures (e.g. standalone numbers, counts, percentages).

    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
     geniş
    -0.94
     غذایی
    -0.88
    かけて
    -0.87
     Paesi
    -0.85
     således
    -0.84
    uomo
    -0.84
    charpe
    -0.84
    愛的
    -0.83
     والت
    -0.82
    諏訪
    -0.82
    POSITIVE LOGITS
    וּ
    1.26
     transparency
    1.04
     whistle
    0.98
     watchdog
    0.97
     environmental
    0.95
    ִּ
    0.94
    Transparency
    0.94
    וֹ
    0.93
    ַּ
    0.92
    CLIENT
    0.92
    Activations Density 0.112%

    No Known Activations