© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 68432
    Prev
    Next
    INDEX
    Explanations

    modes of transport and communication

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

    The neuron activates on prepositions (especially “by” or “via”) that introduce modes of transport (e.g. “traveling by air,” “arrived by sea,” “accessible via water”).

    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
    铪
    -1.63
     durante
    -1.59
     cuatro
    -1.56
    玙
    -1.55
     spécialiste
    -1.52
    用来
    -1.51
     {
    -1.50
    </h1>
    -1.48
     ludzie
    -1.48
    \
    -1.48
    POSITIVE LOGITS
    棄
    1.60
     sedia
    1.60
    Despite
    1.58
     with
    1.55
     These
    1.53
     includes
    1.48
     diretta
    1.48
     ....
    1.47
     Originally
    1.45
     становятся
    1.45
    Activations Density 0.148%

    No Known Activations