© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 72117
    Prev
    Next
    INDEX
    Explanations

    Hon. followed by names or J

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

    The neuron fires on standalone code or configuration keywords and identifiers—things like macro names, build‐script directives, and XML tag names.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     then
    -2.25
     because
    -1.86
     also
    -1.66
    <bos>
    -1.66
    емый
    -1.64
    また
    -1.55
     
    -1.48
     Uniwers
    -1.47
     you
    -1.46
     now
    -1.45
    POSITIVE LOGITS
     the
    2.03
    卧槽
    1.67
     uſ
    1.55
    ⵌ
    1.51
    every
    1.48
     sociala
    1.48
    Sied
    1.47
     Hvis
    1.45
    ًا
    1.45
    various
    1.45
    Activations Density 0.030%

    No Known Activations