© 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. 10-GEMMASCOPE-RES-131K
    7. 102438
    Prev
    Next
    INDEX
    Explanations

    common denominator or Holden

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

    The neuron strongly activates on occurrences of the term “denominator,” especially in the phrase “common denominator.”

    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
    窳
    -2.67
    та
    -2.56
     If
    -2.41
    赂
    -2.30
     ingrained
    -2.30
    しました
    -2.28
    籥
    -2.27
    H
    -2.25
     create
    -2.22
    。】
    -2.22
    POSITIVE LOGITS
     dieß
    2.98
    姈
    2.78
     wth
    2.66
     specif
    2.50
    へ向
    2.45
     洗濯
    2.45
    ↵
    2.44
    0
    2.42
    Ꮴ
    2.39
     BECAUSE
    2.31
    Activations Density 0.003%

    No Known Activations