© 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. 28118
    Prev
    Next
    INDEX
    Explanations

    programming keywords, copyright notices, urls

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

    The neuron fires on code lines that bring in or expose external definitions—e.g. Ruby “require” and “include” (and similar test‐setup keywords like “let”) statements.

    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
     biela
    -1.56
     barcos
    -1.52
     saisi
    -1.52
     tragé
    -1.51
     motoru
    -1.48
    ｮ
    -1.46
     kabát
    -1.46
     dimas
    -1.44
     bicicleta
    -1.44
     histó
    -1.44
    POSITIVE LOGITS
    o
    1.75
    u
    1.66
    d
    1.66
    r
    1.63
     also
    1.57
    e
    1.52
    s
    1.49
    '
    1.48
    g
    1.47
    w
    1.46
    Activations Density 0.045%

    No Known Activations