© 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. 98965
    Prev
    Next
    INDEX
    Explanations

    body tracts and genetics

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

    The neuron primarily activates for the medical/anatomical term "tract" when used in reference to bodily systems or pathways.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron strongly activates on the substring “tract,” i.e. occurrences of the word “tract” in anatomical or biological contexts.

    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
     撮っ
    -2.83
    鲧
    -2.73
     bricola
    -2.63
     GOtt
    -2.56
    注意事
    -2.56
     polie
    -2.55
     ส์
    -2.50
    
    -2.50
     caricatura
    -2.50
     sirena
    -2.50
    POSITIVE LOGITS
    u
    2.86
    D
    2.84
    er
    2.84
    le
    2.83
     '
    2.80
    c
    2.75
    E
    2.75
    ,"
    2.72
     pinpoint
    2.48
     usual
    2.42
    Activations Density 0.004%

    No Known Activations