© 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. Qwen3.5-9B Base
    5. 15-QWENSCOPE-RES-64K
    7. 32484
    Prev
    Next
    INDEX
    Explanations

    gradient and derivative computations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    zinho
    -0.08
    olong
    -0.08
    å®ļ为
    -0.07
    æľ¬èī²
    -0.07
     parado
    -0.07
    elang
    -0.06
     redraw
    -0.06
     pocket
    -0.06
    ÑģÑĤÑĸ
    -0.06
     forcibly
    -0.06
    POSITIVE LOGITS
    enin
    0.08
     Muj
    0.08
    iability
    0.08
    iable
    0.08
    jac
    0.08
    éĻĽ
    0.07
    Incre
    0.07
    scha
    0.07
    ико
    0.07
    ikir
    0.07
    Activations Density 0.009%

    No Known Activations