© 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. 3755
    Prev
    Next
    INDEX
    Explanations

    assistant coach

    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
    éĩį温
    -0.10
    使çĶ¨æĿĥ
    -0.09
    ä¸Ģç¯ĩ
    -0.09
    ä¸Ģç¯ĩæĸĩç«ł
    -0.09
    é¦Ĩ
    -0.08
    ä¸Ģä¸Ŀ
    -0.08
    çĸı
    -0.08
    æİ£
    -0.08
    è¿ĺä¸įæĺ¯
    -0.08
    5
    -0.08
    POSITIVE LOGITS
    ALAR
    0.10
    (extra
    0.09
    RLF
    0.09
    ion
    0.09
    /full
    0.09
    кÑĢа
    0.09
    iliary
    0.09
    rog
    0.09
    æݪ
    0.08
    iliar
    0.08
    Activations Density 0.003%

    No Known Activations