© 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. 34-GEMMASCOPE-RES-131K
    7. 58009
    Prev
    Next
    INDEX
    Explanations

    code structure or variables

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

    This neuron fires on short lowercase identifier tokens in code (e.g. variable names like “buf”, “r”, “esc”, “word”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     with
    -1.13
     in
    -1.01
    ～。
    -0.96
     enhanced
    -0.95
     does
    -0.93
     enhancing
    -0.91
     on
    -0.89
     BytesLike
    -0.88
     just
    -0.88
     enhance
    -0.88
    POSITIVE LOGITS
     Erkrankungen
    1.21
     length
    1.13
     size
    1.10
    為に
    1.05
     последу
    1.03
    BorderWidth
    1.02
    苤
    1.00
    sizeof
    0.99
     "*
    0.99
     '*
    0.98
    Activations Density 0.018%

    No Known Activations