© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 130437
    Prev
    Next
    INDEX
    Explanations

    east of, west of, south of, north of

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

    The neuron activates on numeric measurement tokens—especially distances or coordinates (e.g. decimal numbers indicating how far or where something lies).

    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
    6
    -1.70
     for
    -1.61
     sufficient
    -1.61
    0
    -1.55
     saada
    -1.48
    //
    -1.46
     of
    -1.46
     khawatir
    -1.44
     joten
    -1.42
     koska
    -1.41
    POSITIVE LOGITS
    逦
    1.87
     "'");
    1.83
    ᓚ
    1.70
    ཌ
    1.70
    dding
    1.68
     @"/
    1.62
     ")";
    1.60
     ';
    
    1.55
     "]";
    1.53
     side
    1.52
    Activations Density 0.142%

    No Known Activations