© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8784
    Prev
    Next
    INDEX
    Explanations

    game tokens or specific punctuation

    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
    æİ¥éĢģ
    -0.10
    eson
    -0.10
    çªģåıij
    -0.10
    活泼
    -0.09
    éĺ´è°ĭ
    -0.09
    ç±´
    -0.09
     игÑĢовÑĭе
    -0.09
    çĬ¹
    -0.09
    èī°å·¨
    -0.09
     Army
    -0.09
    POSITIVE LOGITS
    æ¦ľ
    0.12
    ç¼ij
    0.10
    ä¼ij
    0.09
    hek
    0.08
    Asc
    0.08
    éī´å®ļ
    0.08
     guild
    0.08
     Soph
    0.08
    æ¯Ķä¾ĭ
    0.08
    æľĿ
    0.08
    Activations Density 0.100%

    No Known Activations