© 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. 3872
    Prev
    Next
    INDEX
    Explanations

    time to build or complete

    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
    ÑħÑĢан
    -0.10
    ç«ĭåį³
    -0.09
    egis
    -0.09
    é¢ij
    -0.09
    oyer
    -0.09
    æ·±å¤ľ
    -0.09
     submodule
    -0.09
     ÐĴам
    -0.09
    åĢŁéī´
    -0.09
    亿ä¸ĩ
    -0.09
    POSITIVE LOGITS
    æĺİçĻ½
    0.10
    è¾¾åĪ°
    0.10
    éĶ¤
    0.10
    èĦ±ç¦»
    0.09
    çŃĶå¤į
    0.09
    éĩIJ
    0.09
    æIJŀå®ļ
    0.09
    çIJĨä¼ļ
    0.09
    ç¢Ł
    0.08
    æĥ³èµ·æĿ¥
    0.08
    Activations Density 0.028%

    No Known Activations