© 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. 11470
    Prev
    Next
    INDEX
    Explanations

    Redistributions of

    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
    æŃ¤åŁºç¡Ģä¸Ĭ
    -0.13
    åIJİèĢħ
    -0.12
    对åħ¶è¿Ľè¡Į
    -0.11
    äºĮèĢħ
    -0.10
    æĢ»å·¥ä¼ļ
    -0.10
    ä¹Łæĺ¯å¦ĤæѤ
    -0.10
    ä¹Łä¸º
    -0.10
    åĪĻæĺ¯
    -0.10
    ä¹ŁæĹłæ³ķ
    -0.10
    åıijå¸ĥæŶéĹ´
    -0.10
    POSITIVE LOGITS
    <![
    0.09
    duit
    0.09
    åŁĭ
    0.08
    å¤Ł
    0.08
     Dund
    0.08
    º«
    0.08
    ãģ§ãģĤãĤĭãģĵãģ¨
    0.07
    دÙĬد
    0.07
    å±ĤéĿ¢
    0.07
    '),('
    0.07
    Activations Density 0.399%

    No Known Activations