© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 75710
    Prev
    Next
    INDEX
    Explanations

    forms part of

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

    This neuron strongly responds to the multiword phrase “part of” (as in “forms part of” or “forming part of”).

    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
     puramente
    -1.77
     wenigstens
    -1.66
    .
    -1.63
    us
    -1.59
    ままで
    -1.59
     decisivo
    -1.58
     contando
    -1.56
     considerando
    -1.52
    <blockquote>
    -1.50
     zumindest
    -1.49
    POSITIVE LOGITS
     in
    1.92
     лук
    1.68
     et
    1.66
     of
    1.64
     für
    1.61
     the
    1.59
     also
    1.59
    ਸ਼
    1.58
     друго
    1.55
    leitung
    1.55
    Activations Density 0.006%

    No Known Activations