© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 9579
    Prev
    Next
    INDEX
    Explanations

    code structure initialization

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

    The neuron strongly activates on numeric tokens—digits and number literals (including numbers embedded in identifiers) in code.

    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
     habiendo
    -1.19
     verdaderas
    -1.18
     jugó
    -1.13
     mercancías
    -1.10
     sosteniendo
    -1.10
     sanciones
    -1.09
     impresionantes
    -1.09
     murieron
    -1.04
    感じです
    -1.02
    感じが
    -1.02
    POSITIVE LOGITS
     started
    1.01
    说起
    0.95
    Voci
    0.93
    小
    0.92
    success
    0.92
     sucul
    0.91
    外し
    0.90
     luminos
    0.90
    まずは
    0.88
    uktion
    0.88
    Activations Density 0.004%

    No Known Activations