© 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. 129418
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on section and article headings, i.e. the very first token of titles or major segment headers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    diversity of languages

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     after
    -1.28
     on
    -1.28
     at
    -1.24
     as
    -1.16
     in
    -1.16
     or
    -1.13
     without
    -1.00
     when
    -0.98
     if
    -0.96
     JUST
    -0.96
    POSITIVE LOGITS
     alguns
    1.45
     alcuni
    1.38
     alcune
    1.35
    有哪些
    1.33
     algumas
    1.30
     variety
    1.26
     berbagai
    1.22
     principaux
    1.21
    ventory
    1.19
     diversos
    1.16
    Activations Density 0.253%

    No Known Activations