© 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. 58934
    Prev
    Next
    INDEX
    Explanations

    rebuilt and replaced

    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
    å·®çĤ¹
    -0.10
     sÆ°
    -0.10
    ä¸įçĪ±
    -0.10
    大åŁİå¸Ĥ
    -0.09
    åĬŀ好
    -0.09
    è¦ģåģļåĪ°
    -0.09
    .expires
    -0.09
    å·¥ä¸ļä¼ģä¸ļ
    -0.09
    举æŃ¢
    -0.09
    åĨµä¸Ķ
    -0.09
    POSITIVE LOGITS
    åIJİç»Ń
    0.12
    éĩįæĸ°
    0.11
     replaced
    0.11
    éĩįè¿Ķ
    0.11
     replacements
    0.10
    åĨįæĿ¥
    0.10
     replacement
    0.09
    æģ¢å¤į
    0.09
    追究
    0.09
    éļıåIJİ
    0.09
    Activations Density 0.206%

    No Known Activations