© 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. 31428
    Prev
    Next
    INDEX
    Explanations

    originally published/written

    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
    InBackground
    -0.10
    å¹²éĥ¨éĺŁä¼į
    -0.09
    æİĬ
    -0.09
    éĿĴæĺ¥
    -0.09
    梯éĺŁ
    -0.09
    bach
    -0.08
    iban
    -0.08
    èĴ¸æ±½
    -0.08
    çĻ¾å¹´
    -0.08
    ppy
    -0.08
    POSITIVE LOGITS
    Originally
    0.11
     Originally
    0.11
     originally
    0.11
    éĢŀ
    0.09
    çļĦéĥ¨åĪĨ
    0.09
    ocache
    0.09
    为æĬĵ
    0.09
    inese
    0.09
    天涯
    0.09
    ä¸Ĭ课
    0.08
    Activations Density 0.167%

    No Known Activations