© 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. 57491
    Prev
    Next
    INDEX
    Explanations

    color fading or presence

    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
    æĬĬæĪij
    -0.10
    讲课
    -0.09
     Ø´ÙĬ
    -0.09
    æĭ¥
    -0.09
    artic
    -0.08
    åIJ¬å¾Ĺ
    -0.08
    èĮ«èĮ«
    -0.08
    æĶ¶æ¬¾
    -0.08
    åIJĮæľŁ
    -0.08
    èģĭ
    -0.08
    POSITIVE LOGITS
    é«ĺéĵģ
    0.11
    æŁĵ
    0.10
    _ctx
    0.10
    åįŁ
    0.09
    éĿĽ
    0.09
    _sensitive
    0.09
    URES
    0.09
    顺
    0.09
    åIJ¸çĥŁ
    0.09
    åħij
    0.08
    Activations Density 0.134%

    No Known Activations