© 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. 34-GEMMASCOPE-RES-131K
    7. 57493
    Prev
    Next
    INDEX
    Explanations

    boundaries and borders

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

    The neuron is primarily activated by high‐frequency English function words—especially articles like “the” and “a,” and other common stopwords.

    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
    allax
    -0.80
     통해
    -0.75
    ASON
    -0.74
    Destinations
    -0.74
    Ola
    -0.73
     lontano
    -0.73
    扁平
    -0.72
     Commande
    -0.72
    emplos
    -0.72
    dsky
    -0.71
    POSITIVE LOGITS
     boundaries
    6.16
     boundary
    5.56
     Boundaries
    4.63
    boundary
    4.47
    boundaries
    4.34
     borders
    4.22
    Boundary
    4.22
     Boundary
    4.13
    Boundaries
    4.06
     border
    3.81
    Activations Density 0.150%

    No Known Activations