© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36745
    Prev
    Next
    INDEX
    Explanations

    one thing

    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
    éģĹ
    -0.09
    оÑı
    -0.09
    æĦıæĦ¿
    -0.09
     Ara
    -0.08
    éĢĶä¸Ń
    -0.08
    oins
    -0.08
    符åIJĪ
    -0.08
    atre
    -0.08
    odd
    -0.08
    èĪŀ
    -0.08
    POSITIVE LOGITS
    æĪĸå¤ļ
    0.12
    æIJŀ好
    0.11
    å°¤åħ¶æĺ¯
    0.11
    itably
    0.10
    çī¹åĪ«æĺ¯
    0.10
    едÑĮ
    0.10
    ìĶ©
    0.10
    羣æŃ£åģļåĪ°
    0.10
     ấy
    0.09
    atat
    0.09
    Activations Density 0.054%

    No Known Activations