© 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. 61076
    Prev
    Next
    INDEX
    Explanations

    on the other hand

    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
     zwar
    -0.10
    çļĦ主è¦ģ
    -0.09
    azzi
    -0.09
     ÑħоÑĤÑı
    -0.09
    æĶ¯éĥ¨
    -0.09
    æĬī
    -0.09
    åŀ£
    -0.08
     BÃŃ
    -0.08
    èĩ»
    -0.08
    å®Ľå¦Ĥ
    -0.08
    POSITIVE LOGITS
    ä¹Łåı¯èĥ½
    0.10
    æĢĿãģĨ
    0.10
    ä¹Łåı¯ä»¥
    0.10
    ä¹Łæľī
    0.09
    asic
    0.09
    æĺ¯ä¸Ģéĥ¨
    0.09
    è¿Ļä¹Łæĺ¯
    0.09
    pets
    0.09
    æįİ
    0.09
    ä¹Łåı¯
    0.08
    Activations Density 0.004%

    No Known Activations