© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 57028
    Prev
    Next
    INDEX
    Explanations

    PhpStorm, Magick, brush, lab

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    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
    é¹Ĭ
    -0.11
    oning
    -0.09
    æĢľ
    -0.09
    ìĤ¬íķŃ
    -0.09
    åı¯æĢľ
    -0.09
    ÑĩаÑĤ
    -0.09
    تعا
    -0.09
     полиÑĤик
    -0.09
     Atlantis
    -0.09
    ees
    -0.09
    POSITIVE LOGITS
    ol
    0.10
    ruk
    0.09
    ä¸ĭçļĦ
    0.09
     знак
    0.08
    elper
    0.08
    wards
    0.08
    RON
    0.08
    ØŃÙĪ
    0.08
    _gain
    0.08
    chor
    0.08
    Activations Density 0.011%

    No Known Activations