© 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. 33269
    Prev
    Next
    INDEX
    Explanations

    header comments and code declarations

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

    The neuron fires whenever it sees tokens that name low‐level compiler or runtime support libraries or headers (e.g. “CompilerServices,” “defs,” “Expressions”).

    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
     on
    -1.31
    In
    -1.15
    粉絲
    -1.14
     suggest
    -1.13
    One
    -1.11
     find
    -1.10
     classy
    -1.09
    řeba
    -1.08
     are
    -1.07
     One
    -1.07
    POSITIVE LOGITS
    hibit
    1.63
     trám
    1.59
    
    1.50
    ፲
    1.43
    donde
    1.37
    ܊
    1.37
    
    1.36
     entrer
    1.36
    cursions
    1.32
     Después
    1.31
    Activations Density 0.002%

    No Known Activations