© 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. 22508
    Prev
    Next
    INDEX
    Explanations

    as early as

    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
    å¿ĥå¾Ĺä½ĵä¼ļ
    -0.10
    殿
    -0.10
    å¤ĩ
    -0.10
    æĦģ
    -0.09
    è¿ĺæĥ³
    -0.09
    çľĭçľĭ
    -0.09
    Ãłng
    -0.09
    弩
    -0.09
    deo
    -0.09
    é¼İ
    -0.09
    POSITIVE LOGITS
    ä¹İ
    0.10
    룴
    0.09
    为ä¸ŃåĽ½
    0.08
    æľºåĬ¨
    0.08
     fluent
    0.08
     anywhere
    0.08
    äºİ
    0.08
     ëĭ
    0.08
    æ¼Ķ
    0.08
     flush
    0.08
    Activations Density 0.040%

    No Known Activations