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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 30485
    Prev
    Next
    INDEX
    Explanations

    more, news, jobs, post

    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
    foon
    -0.10
    ordinate
    -0.09
    æį»
    -0.09
    é¢ı
    -0.08
    .googlecode
    -0.08
     CRT
    -0.08
    以åıĬåħ¶ä»ĸ
    -0.08
    çİĭåŃIJ
    -0.08
    åĽŀ头
    -0.08
     rev
    -0.08
    POSITIVE LOGITS
    æĹŃ
    0.10
    è¶Ĭ大
    0.10
    æĢģåĬ¿
    0.09
    ä¸Ģæ³¢
    0.09
    ailer
    0.09
    (ib
    0.09
    丼
    0.09
    èĪ·
    0.08
    studio
    0.08
    åı
    0.08
    Activations Density 0.084%

    No Known Activations