© 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. 1922
    Prev
    Next
    INDEX
    Explanations

    ordinal numbers followed by descriptions

    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
    etta
    -0.11
    jÄĻ
    -0.08
    ç¡
    -0.08
     Ñĥни
    -0.08
    itus
    -0.08
    毫æĹłçĸij
    -0.08
     Convenient
    -0.08
    æ®ĵ
    -0.08
    延éķ¿
    -0.08
    ever
    -0.08
    POSITIVE LOGITS
    æļ®
    0.10
     scho
    0.10
    æħij
    0.10
    Collector
    0.09
    ALAR
    0.09
    çĤ¯
    0.09
    éģı
    0.09
    ä¸Ģå¹ķ
    0.08
    æĿ¥æī¾
    0.08
    æľ¬ç½ij
    0.08
    Activations Density 0.011%

    No Known Activations