© 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. 18510
    Prev
    Next
    INDEX
    Explanations

    mania / manic

    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
    绶
    -0.11
    moz
    -0.11
    ahl
    -0.09
    ings
    -0.09
    åĦĨ
    -0.09
    èķ¤
    -0.08
    mund
    -0.08
     táºŃn
    -0.08
    æıIJ
    -0.08
    bourne
    -0.08
    POSITIVE LOGITS
    æĥķ
    0.11
    è·¸
    0.10
    utan
    0.09
    ../../
    0.09
    æīĭå¥Ĺ
    0.09
    èĪ¬çļĦ
    0.09
    aco
    0.09
    ìłģìľ¼ë¡ľ
    0.09
    ç¡ķ
    0.08
    èĢĮåĩº
    0.08
    Activations Density 0.004%

    No Known Activations