© 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. 3025
    Prev
    Next
    INDEX
    Explanations

    call off or play from

    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
    åĿIJçĿĢ
    -0.10
     infer
    -0.09
    ä¸Ģäºĭ
    -0.08
    梳
    -0.08
    çī¢
    -0.08
    ility
    -0.08
    istro
    -0.08
    论述
    -0.08
    moth
    -0.08
    alm
    -0.08
    POSITIVE LOGITS
    åIJ¹
    0.11
    åĽ½æĥħ
    0.10
    éĢĤæŶ
    0.10
    åĴ«
    0.10
    æ··åĩĿ
    0.09
    é£İåIJ¹
    0.09
    èݲèĬ±
    0.09
    èįĶæŀĿ
    0.09
     olds
    0.09
    失误
    0.09
    Activations Density 0.064%

    No Known Activations