© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 109037
    Prev
    Next
    INDEX
    Explanations

    highway exit interchange

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

    The neuron fires strongly on numeric tokens—especially multi‐digit highway, route, or exit numbers in directions or road designations.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    dius
    -0.90
    ట్
    -0.76
    mpagne
    -0.75
    ígenes
    -0.74
    anden
    -0.73
    ,-,
    -0.72
     ἐν
    -0.72
    retmen
    -0.71
    readFileSync
    -0.71
    conto
    -0.70
    POSITIVE LOGITS
     exit
    4.44
     Exit
    4.03
     exits
    3.66
    Exit
    3.56
    exit
    3.52
     EXIT
    3.02
    EXIT
    2.84
     exited
    2.61
    exits
    2.53
     interchange
    2.27
    Activations Density 0.045%

    No Known Activations