© 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. 100273
    Prev
    Next
    INDEX
    Explanations

    visit, walk, traversed

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

    This neuron detects traversal‐related action words—especially forms of “visit” (visit, visitor, visited) and “walk” (Walk, walk, walking) used for walking or traversing data structures.

    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
     ใน
    -1.88
    骉
    -1.74
    齶
    -1.71
     seguintes
    -1.70
     outras
    -1.64
    
    -1.63
     другие
    -1.63
     ;
    -1.61
    つた
    -1.61
    怍
    -1.60
    POSITIVE LOGITS
    </sup>
    2.47
    </b>
    2.25
    𞤢
    2.22
    pecabe
    1.95
    𞤫
    1.95
    躂
    1.88
    Setiap
    1.84
    Kenapa
    1.78
    where
    1.77
    Mengapa
    1.77
    Activations Density 0.007%

    No Known Activations