© 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. 22759
    Prev
    Next
    INDEX
    Explanations

    top of

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

    The neuron fires on words about placing or moving content to the top (or bottom) of a page/view—e.g. “top,” “scroll,” “up,” “page,” and “bottom.”

    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
     a
    -1.98
     those
    -1.80
     TWO
    -1.78
    侮
    -1.65
     both
    -1.63
     our
    -1.55
     ALWAYS
    -1.51
     their
    -1.47
     encantador
    -1.43
    𝑃
    -1.38
    POSITIVE LOGITS
    3
    1.45
    7
    1.39
    スク
    1.33
    必须要
    1.32
     artific
    1.29
    还是要
    1.28
    5
    1.28
    or
    1.27
     Asunción
    1.26
     GARD
    1.22
    Activations Density 0.015%

    No Known Activations