© 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. 21717
    Prev
    Next
    INDEX
    Explanations

    string type declarations

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

    The neuron responds to code tokens that name string‐handling helper functions (e.g. build_arg_string, fmt_doc_string, kwargs_to_strings).

    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
     Pastikan
    -1.70
     vacunas
    -1.66
    lernen
    -1.64
    ケモン
    -1.63
     you
    -1.62
    Ꮟ
    -1.60
     Gunakan
    -1.57
    dengan
    -1.55
    灃
    -1.54
    趿
    -1.54
    POSITIVE LOGITS
    :
    2.06
    -
    2.03
    ↵
    1.83
     to
    1.82
    "
    1.80
    '
    1.71
    A
    1.70
    ?
    1.67
    is
    1.57
    And
    1.57
    Activations Density 0.019%

    No Known Activations