© 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. 30685
    Prev
    Next
    INDEX
    Explanations

    gradient method

    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
    满满çļĦ
    -0.10
    åįģ足
    -0.10
     conc
    -0.09
    conc
    -0.09
    çĤ¹æ»´
    -0.08
    ä¸Ŀ毫
    -0.08
     dar
    -0.08
    åĪĢ
    -0.08
    é²ľ
    -0.08
    çĶļèĩ³
    -0.08
    POSITIVE LOGITS
    åı¦ä¸Ģä½į
    0.10
    èĨ¦
    0.09
    agos
    0.09
    æĪijä¸įçŁ¥éģĵ
    0.09
    åĪ©çī©
    0.09
    noinspection
    0.08
    æĿ¥ä¸įåıĬ
    0.08
    odate
    0.08
    ENO
    0.08
    inand
    0.08
    Activations Density 0.409%

    No Known Activations