© 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. 113143
    Prev
    Next
    INDEX
    Explanations

    undergoing medical procedures

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

    text describing patients undergoing medical procedures or conditions.

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

    The neuron fires most strongly on the central subject or main verb of a clause or sentence—i.e. the key content word that heads the main statement.

    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
     Projektu
    -1.41
     societ
    -1.41
     consid
    -1.41
     teşekkür
    -1.41
     şimdi
    -1.38
    gonna
    -1.37
    ほとん
    -1.36
     müssen
    -1.35
     innumer
    -1.28
     Herramientas
    -1.27
    POSITIVE LOGITS
     with
    1.96
     and
    1.92
     as
    1.87
     all
    1.84
     always
    1.84
     who
    1.82
     on
    1.80
     while
    1.80
     about
    1.78
     every
    1.78
    Activations Density 0.382%

    No Known Activations