© 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. 5671
    Prev
    Next
    INDEX
    Explanations

    noticed that / 've been

    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
    éĤ¬
    -0.10
    rea
    -0.10
    åĴĤ
    -0.09
    çŃ¾åŃĹ
    -0.09
    åŁ¤
    -0.09
    ä¸ºæľŁ
    -0.09
    ä¹IJåĽ¢
    -0.09
    è°
    -0.09
    ä¸Ĭåįĩ
    -0.08
    ä¸Ĭ课
    -0.08
    POSITIVE LOGITS
    å¾Ĥ
    0.11
    owa
    0.09
    ä¸įåħģ许
    0.09
    ä½łä¸į
    0.09
    ä½łä¹Ł
    0.09
    æıIJä¾ĽæĽ´å¤ļ
    0.08
    åĴ
    0.08
    çīĩéĿ¢
    0.08
    erg
    0.08
    èµİ
    0.08
    Activations Density 0.075%

    No Known Activations