© 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. 71883
    Prev
    Next
    INDEX
    Explanations

    assembly, IActionResult, actuated

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

    The neuron fires on code‐style identifiers and symbols—e.g. CamelCase/PascalCase names, function or attribute tokens, and arrow markers (“=>” or “==>”)—that appear in programming or configuration 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_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
     touted
    -2.77
    0
    -2.70
    También
    -2.63
     beinahe
    -2.52
     Gegenstände
    -2.50
    鈑
    -2.47
    ⎙
    -2.38
    錨
    -2.34
     mammoth
    -2.33
     seductive
    -2.33
    POSITIVE LOGITS
    at
    3.09
    you
    2.84
    ad
    2.81
    e
    2.77
    ay
    2.69
    and
    2.48
    `
    2.41
    ist
    2.38
     cambia
    2.38
    b
    2.36
    Activations Density 0.001%

    No Known Activations