© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 40352
    Prev
    Next
    INDEX
    Explanations

    physical contact or obstruction

    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
    Negative Logits
    ä½ĵè´¨
    -0.11
    ä¸Ģ段
    -0.11
    è¿Ļ段
    -0.10
    é¡·
    -0.09
    交ç»Ļ
    -0.09
    æ¸ĹéĢı
    -0.09
    AndWait
    -0.09
    uya
    -0.09
    éĵºè®¾
    -0.08
    äºĨä¸Ģéģį
    -0.08
    POSITIVE LOGITS
    碰
    0.12
    碰æĴŀ
    0.12
    æĮ¡
    0.11
     obstacle
    0.11
    éļľç¢į
    0.10
    pac
    0.10
     obstacles
    0.10
    ç¢į
    0.10
     obstruction
    0.10
    æĴŀ
    0.09
    Activations Density 0.136%

    No Known Activations