© 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. 104294
    Prev
    Next
    INDEX
    Explanations

    t_error characters: ; { } ( ) ; ; ; \; ; ; ; link

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

    semicolons used as delimiters in structured data formats like code, configuration files, and CSV-like data.

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

    This neuron strongly activates on semicolon characters (including LaTeX spacing commands like “\;”), i.e. it’s detecting “;” punctuation.

    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
     now
    -1.46
     because
    -1.40
     Große
    -1.36
     if
    -1.34
     always
    -1.34
     usually
    -1.33
     immensely
    -1.30
     widely
    -1.29
     repeatedly
    -1.28
    ている
    -1.28
    POSITIVE LOGITS
     "..\..\
    1.57
    OrNil
    1.41
    ᕮ
    1.38
    錵
    1.36
    螃
    1.35
    </>
    1.34
     pakai
    1.34
    vä
    1.33
     avec
    1.33
    }$
    1.33
    Activations Density 0.073%

    No Known Activations