© 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. 3480
    Prev
    Next
    INDEX
    Explanations

    general concepts

    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
    <|endoftext|>
    -0.31
    -0.29
     a
    -0.29
     an
    -0.27
     (
    -0.27
     "
    -0.26
     A
    -0.26
     in
    -0.25
     “
    -0.25
     e
    -0.24
    POSITIVE LOGITS
     burge
    0.43
     Girlfriend
    0.39
     giochi
    0.36
     Parenthood
    0.36
     Gratuit
    0.35
     millenn
    0.34
     horr
    0.33
     Marketable
    0.33
     Tradable
    0.32
     Millenn
    0.32
    Activations Density 2.532%

    No Known Activations