© 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. 4585
    Prev
    Next
    INDEX
    Explanations

    remember, know, surprised

    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
    ç¾İ好çļĦ
    -0.11
    åĵªæĢķ
    -0.09
    足足
    -0.09
     kẻ
    -0.09
    _ANGLE
    -0.09
    çļĦç¾İ好
    -0.08
    åŬ
    -0.08
    æ·ĭ
    -0.08
    æĪij们认为
    -0.08
    å§ĭç»Īä¿ĿæĮģ
    -0.08
    POSITIVE LOGITS
    ä¹Łåı¯èĥ½
    0.11
     Conserv
    0.10
    èĤ¯å®ļæĺ¯
    0.10
    èģĶéĢļ
    0.09
    æ´
    0.09
    ä¹Łä¸įæķ¢
    0.09
     conserv
    0.09
     maybe
    0.09
    åı¯èĥ½æĺ¯
    0.09
     conservatism
    0.09
    Activations Density 0.145%

    No Known Activations