© 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. 56923
    Prev
    Next
    INDEX
    Explanations

    common characteristics or features

    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
    å¢Ģ
    -0.11
    ROID
    -0.09
    佯
    -0.09
    çIJ¢
    -0.09
    æ¥Ķ
    -0.09
    arie
    -0.09
    _gid
    -0.08
    TestData
    -0.08
    æĪIJé¾Ļ
    -0.08
    èijº
    -0.08
    POSITIVE LOGITS
    çļĦä¸Ģ项
    0.11
    æī¿
    0.10
    éĩįè¦ģåĨħ容
    0.09
    çļĦä¸Ģ大
    0.09
    çļĦä¸Ģç§į
    0.09
    ä¹ĭä¸Ģ
    0.09
    åħĨ
    0.09
    stalk
    0.09
    ä¸Ĭä¸ĭ
    0.08
    ought
    0.08
    Activations Density 0.213%

    No Known Activations