© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 34008
    Prev
    Next
    INDEX
    Explanations

    code keywords and syntax

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

    This neuron fires on high‐level code syntax markers (e.g. function‐definition keywords like “func” and closing‐tag tokens like “</”).

    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
     by
    -1.91
     in
    -1.49
     of
    -1.45
     with
    -1.39
     as
    -1.38
     to
    -1.30
     which
    -1.27
     at
    -1.20
     on
    -1.15
     because
    -1.15
    POSITIVE LOGITS
    튿
    1.22
    鶫
    1.20
    fantastic
    1.20
     indiquer
    1.19
     três
    1.12
    盉
    1.11
    臜
    1.11
    owano
    1.11
     faciliter
    1.11
     néanmoins
    1.11
    Activations Density 0.035%

    No Known Activations