© 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. 29919
    Prev
    Next
    INDEX
    Explanations

    problem precedent hard specific dangerous

    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
    鬼
    -0.10
     possibly
    -0.10
    æĪĸå¤ļæĪĸå°ij
    -0.09
     ربÙħا
    -0.09
    æŁIJç§įç¨ĭ度
    -0.09
    Runner
    -0.08
    ä¼¼ä¹İæĺ¯
    -0.08
     Possibly
    -0.08
     Atat
    -0.08
    åĪ¨
    -0.08
    POSITIVE LOGITS
    ä¸Ŀ毫
    0.10
    å°±åºĶ该
    0.10
    æķ¬è¯·
    0.09
    éĻĽä¸ĭ
    0.09
    å·²ä¹ħ
    0.09
    åĪ»
    0.09
    ä¿¡èµĸ
    0.08
    ä¸įèĥ½åĨį
    0.08
    已被
    0.08
    æģ¶åĬ£
    0.08
    Activations Density 0.321%

    No Known Activations