© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14489
    Prev
    Next
    INDEX
    Explanations

    descriptor followed by classification

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

    This neuron detects technical mathematical qualifiers and numeric terms such as “fractional,” “integer,” or “modular.”

    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
    があり
    -1.13
     it
    -1.10
    ｽ
    -1.01
     a
    -0.96
    手術
    -0.93
     to
    -0.92
    foundland
    -0.91
    誘惑
    -0.90
    unjungi
    -0.90
    ôs
    -0.89
    POSITIVE LOGITS
     which
    1.30
     sorts
    1.30
     también
    1.16
    際に
    1.13
     ensue
    1.12
    ようだ
    1.12
    heden
    1.09
     jossa
    1.09
     dieťa
    1.08
    しば
    1.07
    Activations Density 0.074%

    No Known Activations