© 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. 2827
    Prev
    Next
    INDEX
    Explanations

    specific terms names

    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
    .ManyToMany
    -0.29
    WithMany
    -0.27
     lineback
    -0.22
    .Dom
    -0.22
    benh
    -0.22
    tsy
    -0.21
    ;!
    -0.21
    otron
    -0.21
    avigate
    -0.21
    URLException
    -0.21
    POSITIVE LOGITS
     Sequelize
    0.21
    è¶Ķ
    0.19
    <decltype
    0.19
    heid
    0.19
    æĵ¤
    0.18
    ě[
    0.18
    æĮ½åĽŀ
    0.18
     testName
    0.18
    常è§Ħ
    0.18
    wie
    0.17
    Activations Density 0.001%

    No Known Activations