© 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. 83966
    Prev
    Next
    INDEX
    Explanations

    or similar

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

    The neuron strongly activates on occurrences of the word “similar” (i.e. markers of similarity/comparison).

    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
    Вита
    -1.55
    睨
    -1.50
    Марина
    -1.43
    いいのか
    -1.43
    Kdo
    -1.40
    栂
    -1.38
    鱨
    -1.38
    Юлия
    -1.38
    躊
    -1.38
    libatkan
    -1.36
    POSITIVE LOGITS
     such
    1.88
     already
    1.47
    ="
    1.45
    安
    1.37
    та
    1.37
    で
    1.34
    '
    1.34
    ly
    1.30
    [
    1.28
     still
    1.23
    Activations Density 0.026%

    No Known Activations