© 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. 35838
    Prev
    Next
    INDEX
    Explanations

    of the

    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
    æıĦ
    -0.11
    è´ŀ
    -0.11
    娱ä¹IJå¹³åı°
    -0.11
    伫
    -0.10
    æŶ髦
    -0.10
    æĿ¥èĩªäºİ
    -0.10
    èIJ¥
    -0.10
    CRET
    -0.10
    让人
    -0.10
    ç¿ķ
    -0.10
    POSITIVE LOGITS
    å·¥ç¨ĭæĬĢæľ¯
    0.08
    座
    0.08
    zb
    0.08
    ç³Ĭ
    0.08
    emma
    0.07
    ot
    0.07
     phenomenon
    0.07
    为导åIJij
    0.07
     Ranger
    0.07
    icks
    0.07
    Activations Density 0.041%

    No Known Activations