© 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. 23472
    Prev
    Next
    INDEX
    Explanations

    numbers followed by points or scores

    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
    å¼Ģåı£
    -0.10
    åIJ¸åħ¥
    -0.09
    èĨı
    -0.09
    èµ°ç§ģ
    -0.09
    人éĢł
    -0.09
    razione
    -0.09
    é¾Ł
    -0.08
    æIJ¶
    -0.08
    æĬ¢æķij
    -0.08
    为主ä½ĵ
    -0.08
    POSITIVE LOGITS
    æľ¬åij¨
    0.09
    åĨĽåĽ¢
    0.09
    inear
    0.08
    onna
    0.08
    owns
    0.08
     Gow
    0.08
    èĮ¹
    0.08
     McCoy
    0.08
    RITE
    0.08
    игÑĢа
    0.08
    Activations Density 0.006%

    No Known Activations