© 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. 10-GEMMASCOPE-RES-131K
    7. 101506
    Prev
    Next
    INDEX
    Explanations

    special characters

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

    question marks in code, particularly for Ruby method names ending with `?` and conditional ternary operators.

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

    The neuron activates on Ruby method names ending with a question mark (i.e. predicate methods).

    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
    us
    -2.20
    Уже
    -2.17
     ostensibly
    -2.17
     unrelenting
    -2.16
    ro
    -2.13
    In
    -2.13
     unparalleled
    -2.11
    ??
    -2.08
     unruly
    -2.06
     =
    -2.00
    POSITIVE LOGITS
     unſ
    2.58
    ientras
    2.52
     และ
    2.48
    鲧
    2.22
     fazendo
    2.17
     delightfully
    2.14
     decidedly
    2.11
    赑
    2.09
     soooo
    2.09
    itsubishi
    2.09
    Activations Density 0.003%

    No Known Activations