© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 13582
    Prev
    Next
    INDEX
    Explanations

    conditional checks and definitions

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

    This neuron activates on C/C++ preprocessor conditional directives (e.g. “#”, “defined”, “ifdef”, numeric literal operands in those 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_10/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
    probably
    -1.77
    both
    -1.68
     zéro
    -1.47
    xically
    -1.41
    before
    -1.37
    through
    -1.37
    yes
    -1.36
     praticamente
    -1.34
    sometimes
    -1.33
     pourront
    -1.32
    POSITIVE LOGITS
     on
    2.36
     to
    2.33
     at
    1.91
     or
    1.78
     that
    1.63
     a
    1.59
     have
    1.57
     new
    1.55
     used
    1.46
     with
    1.45
    Activations Density 0.009%

    No Known Activations