© 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. 22-GEMMASCOPE-RES-131K
    7. 72589
    Prev
    Next
    INDEX
    Explanations

    questions and effects

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

    The neuron is strongly activated by technical biomedical terminology—especially disease names, treatment‐related words (e.g. insulin), and similar clinical‐study jargon.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     händ
    -1.13
    闻言
    -1.10
     prêtre
    -1.05
     пирог
    -1.02
     istimewa
    -1.02
    xxxxxxxxxxxxxxxx
    -0.98
     matiz
    -0.97
     uſ
    -0.97
    DEAN
    -0.96
    ďaka
    -0.95
    POSITIVE LOGITS
     findings
    1.30
     you
    1.30
     your
    1.13
     Ucraina
    1.13
     women
    1.13
     how
    1.13
     choices
    1.10
     know
    1.09
     volu
    1.08
     diversa
    1.08
    Activations Density 0.000%

    No Known Activations