© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36298
    Prev
    Next
    INDEX
    Explanations

    ranking out of

    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
    ç«ĭ
    -0.09
    .dark
    -0.09
    obi
    -0.09
    éĢ¢
    -0.09
    -Mart
    -0.09
     Heller
    -0.08
    ér
    -0.08
    azz
    -0.08
    lew
    -0.08
    ÑģеÑĤ
    -0.08
    POSITIVE LOGITS
     among
    0.16
    among
    0.14
     Among
    0.13
     amongst
    0.12
    Among
    0.12
     ÑģÑĢеди
    0.12
    ä¼Ĺå¤ļ
    0.11
    æ·±æ¸Ĭ
    0.11
     wÅĽród
    0.11
     ×ŀת×ķ×ļ
    0.10
    Activations Density 0.119%

    No Known Activations