© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 6757
    Prev
    Next
    INDEX
    Explanations

    suffix "ers" or proper nouns

    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
    lets
    -0.10
    éľ
    -0.09
    大家åı¯ä»¥
    -0.08
    å®ĥ们
    -0.08
    ä½ľèĢħæľ¬äºº
    -0.08
    被人
    -0.08
    éķ¿çĽ¸
    -0.07
    æľīåºıæİ¨è¿Ľ
    -0.07
    æĿ¡çº¦
    -0.07
    å¾Ģå¹´
    -0.07
    POSITIVE LOGITS
    ä¸Ģ群
    0.10
    hip
    0.10
    imar
    0.09
    heed
    0.09
    upport
    0.09
    HIP
    0.09
    erializer
    0.09
    åĩºåĵģ
    0.09
    èį·
    0.09
    æ¯ı人
    0.09
    Activations Density 0.197%

    No Known Activations