© 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. 18968
    Prev
    Next
    INDEX
    Explanations

    masses of balls

    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
    女çİĭ
    -0.11
    LS
    -0.09
    ä¸Ģå¤ľ
    -0.08
    Cop
    -0.08
    èIJĿ
    -0.08
    uke
    -0.08
    éĢ¸
    -0.08
    è°·
    -0.08
     rocket
    -0.08
    åIJķ
    -0.08
    POSITIVE LOGITS
    ä¸ĢåĽ¢
    0.21
    åłĨ
    0.14
    æĭ³å¤´
    0.13
    çĸĻ
    0.12
    åĽ¢
    0.12
    ä¸Ģ大åłĨ
    0.12
    ä¸įåĪĨ
    0.11
    ä¸Ģçīĩ
    0.11
    åľĺ
    0.11
    æĬ±åĽ¢
    0.11
    Activations Density 0.131%

    No Known Activations