© 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. 29705
    Prev
    Next
    INDEX
    Explanations

    attention and effort

    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
    èı¡
    -0.09
    åłĩ
    -0.09
    ä¸įä¿¡
    -0.09
    ì´ī
    -0.09
     Ferr
    -0.08
     ferr
    -0.08
    omm
    -0.08
     Hick
    -0.08
    å¾ģ
    -0.08
    åįĴ
    -0.08
    POSITIVE LOGITS
    注æĦıåĬĽ
    0.11
    ç«Ń
    0.11
    æīij
    0.10
    å¡Į
    0.10
    ç²¾åĬĽ
    0.10
    æİı
    0.10
    attention
    0.10
    æĬķåħ¥
    0.09
     attention
    0.09
    åħħæ²Ľ
    0.09
    Activations Density 0.031%

    No Known Activations