© 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. 51057
    Prev
    Next
    INDEX
    Explanations

    articles followed by release, statement, email, interview, letter

    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
    åĩºäºİ
    -0.10
    ä¸įæĿ¥
    -0.08
    éĢģåİ»
    -0.08
    èĿĩ
    -0.08
    让æĤ¨
    -0.08
    åīįæıIJæĺ¯
    -0.08
    ä¸Ģå¥Ĺ
    -0.08
    è¾ij
    -0.08
    è·Łä½ł
    -0.08
    \<^
    -0.08
    POSITIVE LOGITS
    ILI
    0.09
    åŁºè°ĥ
    0.09
    Hur
    0.08
    å¤ĸå¢Ļ
    0.08
    togroup
    0.08
    æ¿Ĥ
    0.08
    ücken
    0.08
    ReturnType
    0.08
    ä¸ļåĨħ人士
    0.08
    鬼
    0.08
    Activations Density 0.024%

    No Known Activations