© 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. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-65K
    7. 859
    Prev
    Next
    INDEX
    Explanations

    Canada

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/qwen2.5-7B-it-layer-20-saes/pile/matryoshka/k-100
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     favor
    -0.31
    behavior
    -0.31
     favored
    -0.31
     honored
    -0.29
     favors
    -0.29
     theater
    -0.28
     Center
    -0.27
     CENTER
    -0.27
     theaters
    -0.27
    Behavior
    -0.27
    POSITIVE LOGITS
    .ca
    0.43
    åĬłæĭ¿å¤§
    0.35
     Aboriginal
    0.34
    åĬłæĭ¿
    0.31
     Honour
    0.31
     Saskatchewan
    0.30
    Labour
    0.30
     Winnipeg
    0.29
     Indigenous
    0.29
     Canadians
    0.29
    Activations Density 2.837%

    No Known Activations