© 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. 113909
    Prev
    Next
    INDEX
    Explanations

    tucking, tucked into, Tucker, nestled

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

    The neuron fires on any occurrence of the substring “tuck” (in various forms such as tuck, tucking, tucked, etc.).

    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
     It
    -2.81
    ஂ
    -2.36
     寫
    -2.34
     nocturne
    -2.34
    秝
    -2.31
     spirituale
    -2.31
    椚
    -2.30
    เมื่อ
    -2.25
    ですが
    -2.22
     stihl
    -2.22
    POSITIVE LOGITS
    能在
    2.58
    Lebih
    2.50
    nitř
    2.38
    でて
    2.36
    Meski
    2.34
    Bukan
    2.33
     elke
    2.27
     tackles
    2.25
     ''
    2.23
    Seperti
    2.23
    Activations Density 0.003%

    No Known Activations