© 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. 54184
    Prev
    Next
    INDEX
    Explanations

    units of measurement

    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
    çĦ¶æĺ¯
    -0.09
    éĿ¢åŃIJ
    -0.08
    wy
    -0.08
    idity
    -0.08
    å¾Ĺ好
    -0.08
    ocl
    -0.08
    atu
    -0.08
    oha
    -0.08
    æīĭæİĮ
    -0.07
     SUBSTITUTE
    -0.07
    POSITIVE LOGITS
    èĽĨ
    0.10
    uilt
    0.10
    对ä¸įèµ·
    0.10
    æĿľåħ°
    0.10
    olian
    0.09
    atif
    0.09
    æĴĩ
    0.09
    ormal
    0.09
    аÑħ
    0.09
    atern
    0.09
    Activations Density 0.041%

    No Known Activations