© 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. 53660
    Prev
    Next
    INDEX
    Explanations

    importance of

    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
    ÑĤал
    -0.09
    ä¸įå®Į
    -0.08
    éĥ¨åĪĨåĨħ容
    -0.08
    tem
    -0.08
    æĪĸå¤ļ
    -0.08
    rock
    -0.08
    ç¿Ķ
    -0.08
    é²į
    -0.08
    amage
    -0.07
    Rock
    -0.07
    POSITIVE LOGITS
    åľ¨äºİ
    0.12
    å¸ľ
    0.11
    éĿŀåĩ¡
    0.11
    å°±åľ¨äºİ
    0.10
    immel
    0.10
    æĬĬè¿Ļ个
    0.10
    ä¿Ŀ管
    0.09
    overn
    0.09
    èī¯å¥½
    0.09
    紧迫
    0.09
    Activations Density 0.034%

    No Known Activations