© 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. 16768
    Prev
    Next
    INDEX
    Explanations

    research findings and methods

    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
    ymb
    -0.09
    æĮ½
    -0.09
    åħľ
    -0.09
    çī¹æ´¾
    -0.09
     Belgi
    -0.09
    模èĮĥ
    -0.09
    åĽ½æľī
    -0.09
    á»ĩp
    -0.08
    istan
    -0.08
    夺åıĸ
    -0.08
    POSITIVE LOGITS
    大èĦij
    0.10
    ç¥ŀç»ı
    0.09
    Psych
    0.09
    å¿ĥçIJĨåѦ
    0.09
     EEG
    0.09
    MRI
    0.09
    ç¥ŀç»ıç³»ç»Ł
    0.09
    æ¶İ
    0.08
    è§ħ
    0.08
    Brain
    0.08
    Activations Density 1.142%

    No Known Activations