© 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. 39333
    Prev
    Next
    INDEX
    Explanations

    plural forms after parentheses or slashes

    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
    幺
    -0.09
    峤
    -0.09
    åıĤ
    -0.08
    кÑĢа
    -0.08
    æ·®
    -0.08
    alse
    -0.08
    髯
    -0.08
    楼ä¸Ĭ
    -0.08
    西路
    -0.08
     trem
    -0.08
    POSITIVE LOGITS
    (s
    0.15
    plural
    0.14
    们
    0.14
    åįķä¸Ģ
    0.12
    singular
    0.12
    åĢij
    0.12
    ä¸įæŃ¢
    0.12
    们éĥ½
    0.11
     singular
    0.11
    æĪĸå¤ļ
    0.11
    Activations Density 0.094%

    No Known Activations