© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19610
    Prev
    Next
    INDEX
    Explanations

    ball after

    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
    ë²
    -0.10
    ucks
    -0.09
    ä¹ĭä½ľ
    -0.09
    ingo
    -0.09
    tep
    -0.08
    amel
    -0.08
    ä¸ĢåIJĮ
    -0.08
    Specifier
    -0.08
    ç²¾ç¥ŀ
    -0.08
    们éĥ½
    -0.08
    POSITIVE LOGITS
    éİĸ
    0.09
    æĤŃ
    0.09
     reli
    0.09
    ä»ĸå·²ç»ı
    0.09
    æĬ¥è®°èĢħ
    0.09
    éĢ¡
    0.09
    éĵ®
    0.09
    代表大ä¼ļ
    0.08
    éĩįå¤į
    0.08
    çīĽå¸Ĥ
    0.08
    Activations Density 0.007%

    No Known Activations