© 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. 63009
    Prev
    Next
    INDEX
    Explanations

    special characters following identifiers

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

    The neuron fires on numeric literal tokens (i.e. standalone numbers) in code or mathematical 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_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
     Knights
    -0.69
    numeric
    -0.68
    енье
    -0.67
     war
    -0.66
    padu
    -0.66
     =",
    -0.66
    onStop
    -0.65
    karta
    -0.65
     rever
    -0.65
     Dri
    -0.65
    POSITIVE LOGITS
    ֥
    0.86
    ?[
    0.79
    🔘
    0.78
     capteur
    0.77
     Aarhus
    0.76
     autres
    0.75
     fährt
    0.75
     AllMusic
    0.75
    ?-
    0.75
     chaplain
    0.74
    Activations Density 0.059%

    No Known Activations