© 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. 33503
    Prev
    Next
    INDEX
    Explanations

    technical and scientific terms

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

    The neuron activates on specialized technical or scientific jargon (including acronyms and capitalized domain‐specific terms).

    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
     and
    -1.54
     THEIR
    -1.09
     amd
    -1.05
    🫥
    -1.05
     EVEN
    -1.05
     soooo
    -1.02
     aand
    -1.00
     corect
    -1.00
     tâ
    -0.99
     vivement
    -0.98
    POSITIVE LOGITS
     is
    1.30
    另一方面
    1.30
     can
    1.30
     also
    1.27
     similarly
    1.16
     likewise
    1.15
     will
    1.11
     in
    1.09
     meanwhile
    1.08
     because
    1.05
    Activations Density 0.112%

    No Known Activations