© 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. 48983
    Prev
    Next
    INDEX
    Explanations

    disease, condition, sport, metal, subject, continent

    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
    è¿ĺä¸į
    -0.10
    å´®
    -0.09
    çĵ£
    -0.09
    upp
    -0.09
    æĺ¯ä¸Ģä½į
    -0.08
    æĺ¯ä¸Ģ家
    -0.08
    åįļæĸĩ
    -0.08
    蹩
    -0.08
    ä¸įäºĨ
    -0.08
    å¾ĹçŁ¥
    -0.08
    POSITIVE LOGITS
    à¹īำ
    0.09
    ystal
    0.09
    otec
    0.09
    нÑıÑĤ
    0.09
    ny
    0.09
    æĪĸå¤ļ
    0.09
    çĽijäºĭä¼ļ
    0.09
    erial
    0.09
    太éĺ³åŁİ
    0.08
    OWL
    0.08
    Activations Density 0.149%

    No Known Activations