© 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. 4032
    Prev
    Next
    INDEX
    Explanations

    accepted or taken for granted

    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
    è¾ī
    -0.10
    强迫
    -0.10
    èѦåijĬ
    -0.09
    æĩĭ
    -0.09
    ä¾Ŀ
    -0.09
    ikon
    -0.08
    羣å®ŀ
    -0.08
     кÑĢай
    -0.08
    wear
    -0.08
    æķħæĦı
    -0.08
    POSITIVE LOGITS
    置身
    0.11
    usters
    0.10
    åħ¥åĽ´
    0.10
    uster
    0.09
    .accept
    0.09
    åĢĴéľī
    0.09
    身边
    0.09
     accepted
    0.09
     accept
    0.09
    éģĤ
    0.09
    Activations Density 0.035%

    No Known Activations