© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 129653
    Prev
    Next
    INDEX
    Explanations

    diseases and conditions

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

    citations or references in academic papers denoted by brackets, superscripts, or footnote markers.

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

    The neuron activates on superscripted citation or reference markers (e.g., numbered footnotes) in the text.

    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
     in
    -1.91
     As
    -1.69
     After
    -1.61
     According
    -1.60
     Both
    -1.60
     Make
    -1.48
     A
    -1.48
     At
    -1.45
     But
    -1.45
     Give
    -1.44
    POSITIVE LOGITS
     valioso
    1.99
     Conſ
    1.95
     dát
    1.90
    킁
    1.89
     temáticas
    1.87
     rempliss
    1.83
    幛
    1.82
    ㍑
    1.80
    쫀
    1.80
    읖
    1.79
    Activations Density 0.008%

    No Known Activations