© 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. 113210
    Prev
    Next
    INDEX
    Explanations

    format specification

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

    the word "format" in technical code and configuration files.

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

    The neuron fires on instances of the field name “format.”

    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
    を中心
    -2.30
    蹰
    -2.30
    selbe
    -2.30
     legislación
    -2.23
     Сайт
    -2.20
     contribución
    -2.20
     réfugiés
    -2.16
     cojín
    -2.13
    schermata
    -2.11
    蟥
    -2.11
    POSITIVE LOGITS
    ?
    3.06
    a
    2.98
    4
    2.91
    -
    2.72
    To
    2.63
    0
    2.56
    3
    2.53
    ”
    2.53
    1
    2.42
    9
    2.34
    Activations Density 0.018%

    No Known Activations