© 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. 53808
    Prev
    Next
    INDEX
    Explanations

    one or multiple copies

    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
    èĥĮåIJİ
    -0.10
    æŃ¥ä¼IJ
    -0.10
    æľįåĬ¡åĮº
    -0.09
    ç»Ĭ
    -0.09
    -IN
    -0.08
    æĸĩåĮĸåĴĮ
    -0.08
    èį¤
    -0.08
    è¿ĽæŃ¥
    -0.08
    ë°´
    -0.08
    éŨå¤ĸ
    -0.08
    POSITIVE LOGITS
     copies
    0.21
    copies
    0.19
    åī¯æľ¬
    0.18
     copy
    0.16
     Copies
    0.16
    æĭ·
    0.16
    opies
    0.14
    å¤įåį°ä»¶
    0.14
     instances
    0.14
    instance
    0.13
    Activations Density 0.217%

    No Known Activations