© 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. 48536
    Prev
    Next
    INDEX
    Explanations

    behaviour through or file using

    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
    æĪªæŃ¢
    -0.10
    æĪª
    -0.10
    æŀķ头
    -0.09
    ä¿®
    -0.09
    æĪªåĽ¾
    -0.09
    é¹ĥ
    -0.09
    ãģĮãĤī
    -0.09
    黳
    -0.09
    enu
    -0.09
    ën
    -0.09
    POSITIVE LOGITS
    åĸľå¥½
    0.13
    åģı好
    0.09
     Ashton
    0.08
    æ»ŀåIJİ
    0.08
     Rodgers
    0.08
    ä¸ŃéĢĶ
    0.08
     Savage
    0.08
    çĶ¨é¤IJ
    0.08
    忽
    0.08
    åįĹæ²Ļ
    0.08
    Activations Density 0.056%

    No Known Activations