© 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. 18313
    Prev
    Next
    INDEX
    Explanations

    za Shinwari

    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
    çĨ
    -0.09
    .jd
    -0.09
     меÑģÑĤ
    -0.08
     harassing
    -0.08
     Ms
    -0.08
    æĸ°å¼Ģ
    -0.08
    ä¸ļ主
    -0.08
    lazy
    -0.08
    åĪ·æĸ°
    -0.08
     laz
    -0.08
    POSITIVE LOGITS
    çĪ±äºº
    0.14
    æī¿åıĹ
    0.12
    æĦŁæĥħ
    0.11
    çĪ±æĥħ
    0.10
    ä¸ĢçĽ´ä»¥æĿ¥
    0.10
    伤å¿ĥ
    0.10
    çļĦæĦŁæĥħ
    0.10
    ä¸įåĨįæĺ¯
    0.10
    ä»ĬçĶŁ
    0.10
    æľĢéĩįè¦ģ
    0.10
    Activations Density 0.050%

    No Known Activations