© 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. 22-GEMMASCOPE-RES-131K
    7. 26573
    Prev
    Next
    INDEX
    Explanations

    category names XML namespaces

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

    The neuron detects numeric literals (especially multi-digit and decimal version numbers) in code or text.

    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
     in
    -2.03
     because
    -2.00
     supremely
    -1.89
     one
    -1.85
     ostensibly
    -1.73
     estekak
    -1.73
     both
    -1.72
     undeniably
    -1.72
     or
    -1.68
     TWO
    -1.68
    POSITIVE LOGITS
     alcune
    1.97
     Dinas
    1.81
     .........
    1.67
     .......
    1.66
     ........
    1.64
     mediocre
    1.63
     ..........
    1.59
     ___________
    1.59
     verschillende
    1.59
     luscious
    1.55
    Activations Density 0.044%

    No Known Activations