© 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. 31157
    Prev
    Next
    INDEX
    Explanations

    Jaime Ignacio, Lettuce Salad, Mitromorpha

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

    The neuron fires on tokens that appear in “declaration” or “metadata‐header” contexts—things like #include/define/using/import/package lines, license or dependency keys, code‐structure keywords, and similar top‐level definitions.

    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
     ast
    -0.73
    瀾
    -0.73
     é
    -0.73
     chop
    -0.73
     ane
    -0.72
     cada
    -0.71
     tap
    -0.70
     с
    -0.70
     Rhode
    -0.69
     spice
    -0.68
    POSITIVE LOGITS
     Features
    1.12
    Feature
    1.08
    Features
    1.05
     features
    1.04
     feature
    1.02
    feature
    1.02
     FEATURES
    1.02
     Feature
    1.02
    features
    0.98
    FEATURES
    0.94
    Activations Density 0.007%

    No Known Activations