© 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. 10-GEMMASCOPE-RES-131K
    7. 124645
    Prev
    Next
    INDEX
    Explanations

    technical documentation links

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

    The neuron activates on the hyphen separators and the surrounding segments in hyphenated identifiers (e.g. in file names, link slugs, or markdown link labels).

    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
     on
    -2.17
     in
    -2.14
     for
    -1.93
     not
    -1.84
     or
    -1.80
    >
    -1.79
     when
    -1.77
     out
    -1.76
     without
    -1.71
     during
    -1.61
    POSITIVE LOGITS
     célè
    1.80
     régal
    1.80
     diriger
    1.77
     rafra
    1.76
     renou
    1.74
     réinitial
    1.73
     éte
    1.71
     favoriser
    1.71
     rafraî
    1.71
     poë
    1.68
    Activations Density 0.256%

    No Known Activations