© 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. 39407
    Prev
    Next
    INDEX
    Explanations

    to modify

    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
    ĸ
    -0.10
    éħ¥
    -0.10
    åľ¨ä¸Ģèµ·
    -0.09
    å®Į好
    -0.09
    è®®
    -0.09
    éĥ¨ä½į
    -0.09
    çł´æįŁ
    -0.09
    鼶ç¢İ
    -0.08
    å¾Īéķ¿
    -0.08
    vocab
    -0.08
    POSITIVE LOGITS
    使ä¹ĭ
    0.11
    å°ĨåĨĽ
    0.10
    使åħ¶
    0.09
    añ
    0.09
    åĬłåħ¥äºĨ
    0.09
    åĮħåIJ«äºĨ
    0.09
     include
    0.09
    纳åħ¥
    0.08
     emphasis
    0.08
    æĽ´èĥ½
    0.08
    Activations Density 0.144%

    No Known Activations