© 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. 60802
    Prev
    Next
    INDEX
    Explanations

    if condition is

    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
     Reeves
    -0.09
    /apple
    -0.08
     Vy
    -0.07
    ESS
    -0.07
     wastes
    -0.07
    é«¡
    -0.07
    对æѤ
    -0.07
    ìĸij
    -0.07
    è¿Ļåıª
    -0.07
    .online
    -0.07
    POSITIVE LOGITS
    èĢĮæĺ¯
    0.11
    bung
    0.10
    ä¿Ĺ称
    0.09
    afia
    0.09
    æīĢ以æĪij
    0.09
    ucher
    0.09
    åĸĿ
    0.09
    ä»»ä½ķå½¢å¼ı
    0.09
    emu
    0.09
    ä»İèĢĮ使
    0.09
    Activations Density 1.369%

    No Known Activations