© 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. 15927
    Prev
    Next
    INDEX
    Explanations

    specific technical terms

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

    The neuron fires most strongly on words that occur repeatedly in the same document, i.e. it detects locally high term‐frequency tokens.

    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
     it
    -1.26
     in
    -1.11
    Altura
    -0.93
     explorar
    -0.92
     INTERESAR
    -0.91
     provide
    -0.91
     as
    -0.89
    ēl
    -0.88
     Malcolm
    -0.86
    瑠
    -0.85
    POSITIVE LOGITS
     your
    1.23
     Espec
    0.93
     nimmer
    0.91
     konkrét
    0.91
    cedures
    0.86
     می‌گو
    0.85
     استیل
    0.85
    garis
    0.85
     producenta
    0.84
     SPECIFIC
    0.84
    Activations Density 0.090%

    No Known Activations