© 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. 10-GEMMASCOPE-RES-131K
    7. 102768
    Prev
    Next
    INDEX
    Explanations

    software descriptions

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

    the word "description" followed by a colon and opening quotation mark in JSON or structured data formats.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron activates strongly on the quoted text values of JSON “description” fields.

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     hermosas
    -1.81
     begeistert
    -1.55
     Añade
    -1.51
     licencias
    -1.46
     cocinas
    -1.41
     Präsidenten
    -1.40
     ardından
    -1.39
     -¿
    -1.38
     maravill
    -1.38
     respective
    -1.37
    POSITIVE LOGITS
     is
    1.63
    tral
    1.62
    这
    1.55
     ఈ
    1.45
    这个
    1.42
     mignon
    1.42
    they
    1.41
     enää
    1.41
     essendo
    1.40
    meester
    1.38
    Activations Density 0.039%

    No Known Activations