© 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. 888
    Prev
    Next
    INDEX
    Explanations

    preference default automatically

    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
    讫
    -0.11
    çĶµè§Ĩæľº
    -0.11
    åŃĺåľ¨çĿĢ
    -0.11
    åģļä¸įåĪ°
    -0.10
    ç»ĦåĽ¢
    -0.10
    ä¸įåı¯éģ¿åħį
    -0.10
    åħ·æľīèī¯å¥½
    -0.09
    æĮĩæłĩ
    -0.09
    è¶ħæłĩ
    -0.09
     amen
    -0.09
    POSITIVE LOGITS
     prefer
    0.11
    åĭ¿
    0.09
     default
    0.09
     Prefer
    0.08
    idel
    0.08
    é¦ĸ
    0.08
     preference
    0.08
     automatically
    0.08
    é»ĺ认
    0.08
    Preference
    0.08
    Activations Density 0.624%

    No Known Activations