© 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. 60849
    Prev
    Next
    INDEX
    Explanations

    search online, service, function, by

    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
    (expect
    -0.08
    æ¹į
    -0.08
    пи
    -0.08
     hod
    -0.08
    使
    -0.07
    Hack
    -0.07
    èĩ´
    -0.07
    潺
    -0.07
    å¼Ģè¾Ł
    -0.07
    éĹ
    -0.07
    POSITIVE LOGITS
    èĻ±
    0.11
    ability
    0.11
    æĥŃ
    0.10
    æIJľæķij
    0.10
    马路
    0.09
    leck
    0.09
    ç©¿æIJŃ
    0.09
    æĮ¤
    0.09
    isz
    0.09
    ufen
    0.09
    Activations Density 0.065%

    No Known Activations