© 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. 32447
    Prev
    Next
    INDEX
    Explanations

    any questions or suggestions

    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
    åĽŀäºĭ
    -0.11
    ä¸ĭåįĬ
    -0.09
    æľīå¾Īå¤ļç§į
    -0.09
    çĸµ
    -0.09
    æī¿æĭħ责任
    -0.09
    饺
    -0.09
    ICY
    -0.09
    å¾Īåı¯èĥ½
    -0.08
    çļĦä¸Ģç§į
    -0.08
     Andersen
    -0.08
    POSITIVE LOGITS
     neat
    0.08
    ä¸į好
    0.08
    说æ³ķ
    0.08
     Zug
    0.08
     heavy
    0.08
     seri
    0.08
     spare
    0.07
    å¤Ļ
    0.07
    ç»ıåħ¸
    0.07
     loose
    0.07
    Activations Density 0.005%

    No Known Activations