© 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. 992
    Prev
    Next
    INDEX
    Explanations

    to use, anyone, build, appeal, provide

    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
    æĦŁè§īèĩªå·±
    -0.10
    awy
    -0.10
    oriously
    -0.09
    è®°å¾Ĺ
    -0.09
    åĪĽå§ĭ
    -0.09
    åħ»æĪIJ
    -0.09
    ç©°
    -0.09
    éħ±æ²¹
    -0.09
    åĿıäºĨ
    -0.08
    æĭ¦
    -0.08
    POSITIVE LOGITS
    å¼Ģå±ķ
    0.11
    ä»İäºĭ
    0.09
    ç»ıèIJ¥
    0.09
    使çĶ¨
    0.09
    举åĬŀ
    0.09
    uba
    0.09
    ä¸ĸçķĮä¸Ĭ
    0.08
    代表æĢ§
    0.08
    使çĶ¨çļĦ
    0.08
     Thorn
    0.08
    Activations Density 0.047%

    No Known Activations