© 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. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-65K
    7. 362
    Prev
    Next
    INDEX
    Explanations

    being or having

    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
     discrepan
    -0.35
     busty
    -0.34
    :animated
    -0.30
    disposing
    -0.30
    UniversitÃł
    -0.29
    nodeValue
    -0.26
    disposed
    -0.26
     pérd
    -0.26
     órgão
    -0.26
     UClass
    -0.26
    POSITIVE LOGITS
     being
    0.51
    being
    0.40
     having
    0.39
    Being
    0.34
     Being
    0.34
     making
    0.31
     using
    0.30
     getting
    0.30
    -being
    0.28
    having
    0.27
    Activations Density 4.044%

    No Known Activations