© 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. 10-GEMMASCOPE-RES-131K
    7. 40010
    Prev
    Next
    INDEX
    Explanations

    country, lat, currency, role

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

    The neuron primarily activates on programming identifiers—tokens that look like variable, function, class, or constant names (especially those containing uppercase or mixed-case letters).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     with
    -2.42
     for
    -2.08
     on
    -1.99
     at
    -1.91
     does
    -1.87
     did
    -1.86
     even
    -1.84
     which
    -1.76
     more
    -1.67
     when
    -1.63
    POSITIVE LOGITS
    裢
    1.87
    謦
    1.80
    recently
    1.78
    킁
    1.75
    튿
    1.73
     empêcher
    1.73
    すっ
    1.73
    {
    1.71
    represents
    1.71
     wszyscy
    1.71
    Activations Density 0.248%

    No Known Activations