© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20064
    Prev
    Next
    INDEX
    Explanations

    quantitative analysis and prediction

    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
    åģļ大
    -0.09
    Jam
    -0.08
    spar
    -0.08
    æ±Ĺæ°´
    -0.08
     Jam
    -0.08
     incumbent
    -0.08
    ä¹°åįķ
    -0.08
    دÙĨÙĬ
    -0.08
    æĸ¹æĸ¹éĿ¢éĿ¢
    -0.08
    urger
    -0.08
    POSITIVE LOGITS
    ä¼¼çļĦ
    0.08
    æĹłåĬ©
    0.08
    __),
    0.08
    LEMENT
    0.08
    æľīä¿¡å¿ĥ
    0.08
    辦æ³ķ
    0.08
    çĶµèĦij
    0.08
    å¼Ħ
    0.08
    tlement
    0.08
    æª
    0.08
    Activations Density 1.229%

    No Known Activations