© 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. 58650
    Prev
    Next
    INDEX
    Explanations

    being + adjective/noun

    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
    çķ´
    -0.11
    åĬ²
    -0.09
    omen
    -0.09
    more
    -0.08
     contend
    -0.08
    å¢ĥå¤ĸ
    -0.08
    à¹ģà¸ģ
    -0.08
    å¹²èĦĨ
    -0.08
    çĶŁåĬ¨
    -0.08
    çķ¥
    -0.08
    POSITIVE LOGITS
    身为
    0.10
    åľ°å¤Ħ
    0.10
    å·²ç»ıæľī
    0.09
    ä¸ĢåIJij
    0.09
    ä½łæĺ¯
    0.09
    å«Ĥ
    0.09
    æ¯ķç«Łæĺ¯
    0.09
     знаÑİ
    0.09
    uhl
    0.09
    ä¹ŁçŁ¥éģĵ
    0.09
    Activations Density 0.025%

    No Known Activations