© 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. 57801
    Prev
    Next
    INDEX
    Explanations

    labels and classification

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

    The neuron strongly activates on Python code identifiers and keywords (e.g. “def,” “self,” variable names like “x,” “y,” “classes,” “predict,” “return”), effectively detecting when the text is inside a Python method definition or call.

    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
     Inputs
    -0.97
    romptu
    -0.91
     Betten
    -0.83
    inputs
    -0.77
     Calculations
    -0.76
     trưởng
    -0.73
     요
    -0.73
     个性
    -0.72
    fino
    -0.72
    curkan
    -0.72
    POSITIVE LOGITS
     labels
    1.13
     integer
    1.02
     categorical
    0.97
     labelling
    0.93
    binary
    0.90
    argmax
    0.89
     binary
    0.89
    label
    0.89
    LABEL
    0.89
    Labels
    0.87
    Activations Density 0.012%

    No Known Activations