© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-CHAT-65K
    7. 46935
    Prev
    Next
    INDEX
    Explanations

    failing performance

    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/lmsys/matryoshka/k-100
    Prompts (Dashboard)
    16,384 prompts, 192 tokens each
    Dataset (Dashboard)
    chanind/lmsys-chat-1m-qwen-2.5
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     kut
    -0.22
     înt
    -0.21
    ấy
    -0.21
     Åŀa
    -0.20
    .§
    -0.20
    quo
    -0.20
    lickr
    -0.20
     Crush
    -0.20
     Dabei
    -0.20
     Ipsum
    -0.20
    POSITIVE LOGITS
     tslint
    0.22
    ä¸ĸçķĮ级
    0.20
    (strpos
    0.20
    иде
    0.19
    太快
    0.19
    getDescription
    0.19
    OMEM
    0.18
     the
    0.18
    ledged
    0.18
    *>*
    0.18
    Activations Density 0.001%

    No Known Activations