© 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. 22-GEMMASCOPE-RES-131K
    7. 15412
    Prev
    Next
    INDEX
    Explanations

    punctuation and symbols

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

    This neuron activates on snippets of programming code—especially numeric literals (e.g. version numbers), code punctuation, and include/pragma‐style tokens.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    Unmanaged
    -1.16
    了一会儿
    -1.12
     mesmerizing
    -1.08
    ccional
    -1.04
     intric
    -1.02
     astounding
    -1.01
     mediocre
    -1.01
    rii
    -1.00
    FINISH
    -1.00
     horrid
    -0.98
    POSITIVE LOGITS
     that
    1.39
     before
    1.33
     in
    1.30
     what
    1.28
     it
    1.23
     where
    1.13
    With
    1.13
     allows
    1.13
     showed
    1.12
     through
    1.09
    Activations Density 0.004%

    No Known Activations