© 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. 40361
    Prev
    Next
    INDEX
    Explanations

    code block beginnings

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

    The neuron strongly activates on numeric literals (numbers) in code or comments.

    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
     has
    -2.05
     one
    -1.73
     at
    -1.66
     stating
    -1.63
     which
    -1.62
     all
    -1.59
     for
    -1.59
     throughout
    -1.56
     through
    -1.56
     to
    -1.55
    POSITIVE LOGITS
     réduire
    1.62
     dreary
    1.61
     quintessential
    1.60
     supprimer
    1.58
     magnific
    1.55
     arrêter
    1.55
     shimmering
    1.54
     exhilarating
    1.54
     refroid
    1.53
     colossal
    1.53
    Activations Density 0.002%

    No Known Activations