© 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. 19635
    Prev
    Next
    INDEX
    Explanations

    hand, palmar

    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
    åĪ·åį¡
    -0.09
    åŃ¦ä¹łè´¯å½»
    -0.09
    åĸĨ
    -0.08
    åĢļ
    -0.08
    è¿Ľåħ¥åĪ°
    -0.08
    åIJ¦
    -0.08
    è¶´åľ¨
    -0.08
    èĥĦ
    -0.08
    åĸī
    -0.08
    reen
    -0.08
    POSITIVE LOGITS
    æĻ®éĢļ
    0.09
    æīĭæİĮ
    0.09
    éĿ¢ç²ī
    0.09
    wap
    0.09
    è´Łæĭħ
    0.09
    hab
    0.09
    åĿ³
    0.09
    éĥ¤
    0.08
    haul
    0.08
    ä¸Ģå¼ł
    0.08
    Activations Density 0.030%

    No Known Activations