© 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. 63025
    Prev
    Next
    INDEX
    Explanations

    new or next followed by a noun

    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
    æ¯Ķäºļ
    -0.10
    æ¥Ķ
    -0.10
    çĬ¹
    -0.10
    ä¸ĭä¸Ģ代
    -0.10
    æļ¾
    -0.10
    éĽĨåĽ¢èĤ¡ä»½
    -0.09
    åħģ
    -0.09
    çĶŁäº§è®¾å¤ĩ
    -0.09
    æĢĤ
    -0.09
    çĮ·
    -0.09
    POSITIVE LOGITS
    头æĿ¡
    0.09
    åħ¨éķĩ
    0.09
    éķĩæĶ¿åºľ
    0.09
    hom
    0.09
    åįĩéĻį
    0.09
    ion
    0.08
    常æĢģ
    0.08
    /manual
    0.08
     litter
    0.08
    elo
    0.08
    Activations Density 0.035%

    No Known Activations