© 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. 64079
    Prev
    Next
    INDEX
    Explanations

    change up or shake up

    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
    çĶµç½ij
    -0.09
    ires
    -0.08
    æĹłå£°
    -0.08
    çł¥
    -0.08
    æ¸
    -0.08
    æ¶Ķ
    -0.08
    afen
    -0.08
    OUS
    -0.08
    /grid
    -0.08
    ovan
    -0.08
    POSITIVE LOGITS
    åĬ²
    0.11
    æį¢ä¸ª
    0.11
    åľ¨æ·±åľ³
    0.10
    cake
    0.10
    æĢ¦
    0.10
    活泼
    0.09
    è¹´
    0.09
    èĬ±æł·
    0.09
    æľºåĬ¨
    0.09
    åĨħ饰
    0.08
    Activations Density 0.037%

    No Known Activations