© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 11185
    Prev
    Next
    INDEX
    Explanations

    code after punctuation

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

    the period character following "Mvc" in code (especially in namespace and import statements).

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

    The neuron is keying on statement terminators (i.e. semicolons) in C-style code.

    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
     =
    -2.14
     of
    -2.13
    for
    -2.11
     –
    -1.91
     ‘‘
    -1.89
     a
    -1.88
    e
    -1.84
    ?
    -1.84
    de
    -1.82
     —
    -1.73
    POSITIVE LOGITS
    俛
    1.95
     osoba
    1.94
    ﾁ
    1.80
    棡
    1.77
     ſta
    1.75
    และ
    1.74
     pousse
    1.70
     odpad
    1.70
     acabar
    1.70
    силання
    1.69
    Activations Density 0.001%

    No Known Activations