© 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. 15967
    Prev
    Next
    INDEX
    Explanations

    code structures

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

    The neuron strongly activates on programming‐language structural markers (AST node labels or code section tags), i.e. tokens that name or delimit major code constructs.

    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
    nés
    -0.99
     REALLY
    -0.99
    GGI
    -0.94
    ACHT
    -0.94
     "*
    -0.92
    ISMS
    -0.91
    OMES
    -0.91
     powi
    -0.91
     urat
    -0.91
     Dinas
    -0.89
    POSITIVE LOGITS
     another
    1.48
    Another
    1.26
     also
    1.19
     також
    1.15
    While
    1.13
    During
    1.11
     while
    1.09
    同じく
    1.08
     similarly
    1.06
     this
    1.03
    Activations Density 0.001%

    No Known Activations