© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-65K
    7. 476
    Prev
    Next
    INDEX
    Explanations

    often, sometimes, usually

    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
    åIJįåĪĹåīį
    -0.30
    @student
    -0.28
    **)&
    -0.27
    ï¼ģ",
    -0.27
    âļĹ
    -0.26
     subparagraph
    -0.26
    :numel
    -0.26
     **)&
    -0.26
    =findViewById
    -0.26
    .omg
    -0.26
    POSITIVE LOGITS
    å¾Ģå¾Ģ
    0.73
     often
    0.67
    å¾Ģå¾Ģæĺ¯
    0.59
     sometimes
    0.59
    often
    0.54
    å¾Ģå¾Ģä¼ļ
    0.54
     usually
    0.53
     Often
    0.51
    éĢļ常
    0.50
    æľīæŶ
    0.48
    Activations Density 10.293%

    No Known Activations