© 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. 3039
    Prev
    Next
    INDEX
    Explanations

    who/whom/whose

    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
     pä
    -0.23
    embros
    -0.23
    chluss
    -0.21
    obierno
    -0.21
    ntity
    -0.21
    ecom
    -0.20
    gebn
    -0.20
    "<?
    -0.20
    fällig
    -0.20
    .AutoScaleMode
    -0.19
    POSITIVE LOGITS
     who
    0.28
     whom
    0.26
    åIJĮæĹ¶ä¹Łæĺ¯
    0.25
    who
    0.25
     whose
    0.24
     quien
    0.20
     fame
    0.20
    öm
    0.20
     ""}↵
    0.19
    whose
    0.19
    Activations Density 0.918%

    No Known Activations