© 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. 1677
    Prev
    Next
    INDEX
    Explanations

    do what we can

    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
    ä¸ĢèĦ¸
    -0.10
    ÑĻ
    -0.10
    å±Ī
    -0.10
    arium
    -0.09
    èĴ
    -0.09
    ä¸Ģ头
    -0.09
    对æ¯Ķ
    -0.09
    åĹ·
    -0.09
    æľįåĬ¡ä½ĵç³»
    -0.09
    æĶ¹éĿ©åıijå±ķ
    -0.09
    POSITIVE LOGITS
    åģļçļĦäºĭæĥħ
    0.12
    åİ»åģļ
    0.10
    åģļçļĦäºĭ
    0.10
    åģļçļĦ
    0.10
    æĵįä½ľ
    0.09
    åģļå¾Ĺ
    0.09
    åĬŀçIJĨ
    0.09
    æ´»åĭķ
    0.09
     done
    0.09
    æ´»åĬ¨
    0.09
    Activations Density 0.125%

    No Known Activations