© 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. 10877
    Prev
    Next
    INDEX
    Explanations

    code identifiers and punctuation

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

    variable or identifier names in code that consist of multiple words joined by hyphens, underscores, or camelCase.

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

    The neuron activates on numeric literals (integer or decimal constants) scattered through code or style definitions.

    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
    櫺
    -1.49
     of
    -1.38
     escalier
    -1.35
    vamp
    -1.34
     wasnt
    -1.30
    êt
    -1.27
    coloriage
    -1.25
     specif
    -1.23
     atunci
    -1.23
     Bäume
    -1.23
    POSITIVE LOGITS
    What
    1.84
     It
    1.70
    Some
    1.65
     two
    1.59
    Which
    1.57
    Another
    1.55
     Some
    1.53
     nawet
    1.47
     даже
    1.46
     But
    1.41
    Activations Density 0.144%

    No Known Activations