© 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. 714
    Prev
    Next
    INDEX
    Explanations

    type

    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
    variably
    -0.24
    chedulers
    -0.22
    abcdefghijklmnop
    -0.21
    ptune
    -0.19
    erreur
    -0.19
     ÙĦÙħاذا
    -0.18
    itled
    -0.18
    storybook
    -0.18
     equivalents
    -0.18
    æĤĴ
    -0.18
    POSITIVE LOGITS
     type
    0.52
    ç±»åŀĭ
    0.46
     location
    0.37
    .type
    0.37
    type
    0.36
     age
    0.36
    åľ°çĤ¹
    0.35
     types
    0.35
    ç±»åŀĭçļĦ
    0.35
    ç§įç±»
    0.33
    Activations Density 3.574%

    No Known Activations