© 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. 19631
    Prev
    Next
    INDEX
    Explanations

    evaluate factors or reasons to justify

    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
    çļĦä¹łæĥ¯
    -0.09
    ä¹łæĥ¯
    -0.08
    赶
    -0.08
    åıĪæĺ¯
    -0.08
     habit
    -0.08
    æľĿ
    -0.08
    åŁł
    -0.08
    å¾ĹçĽĬäºİ
    -0.08
    æĬķæľº
    -0.07
    æľºä¼ļ
    -0.07
    POSITIVE LOGITS
    æł¹æį®ä¸įåIJĮ
    0.11
    æģ»
    0.10
    ä¾Ľæ±Ĥ
    0.09
    åī¥å¤º
    0.09
    åľ¨åľ°ä¸Ĭ
    0.09
    æ³ķå¾ĭè§Ħå®ļ
    0.09
    ä¾µçĬ¯
    0.09
    ä¹ŁåºĶ该
    0.09
    BLE
    0.09
    责令
    0.08
    Activations Density 0.067%

    No Known Activations