© 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. 6053
    Prev
    Next
    INDEX
    Explanations

    wisdom

    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
    aways
    -0.10
    away
    -0.09
    bies
    -0.09
    zeit
    -0.09
    æ±
    -0.09
    ä¸įä¸Ĭ
    -0.09
    ivate
    -0.08
    æĪijåĸľæ¬¢
    -0.08
    athon
    -0.08
    üns
    -0.08
    POSITIVE LOGITS
    fully
    0.11
    acre
    0.10
    FUL
    0.10
    0.10
    女ç¥ŀ
    0.09
    айд
    0.09
    RuntimeException
    0.09
    anca
    0.09
    ierz
    0.09
     ÑĤоÑĩно
    0.08
    Activations Density 0.020%

    No Known Activations