© 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. 58668
    Prev
    Next
    INDEX
    Explanations

    3D parallax

    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
    çĻ¾èĬ±
    -0.11
    积水
    -0.09
    ä¸ŃæĸŃ
    -0.09
     CEL
    -0.09
    å¾ĹåĪĨ
    -0.09
    é¾
    -0.08
     Nail
    -0.08
    èĥİ
    -0.08
    _FIN
    -0.08
     Ker
    -0.08
    POSITIVE LOGITS
    è¶´åľ¨
    0.12
    åıĮçľ¼
    0.11
    çľ¼ä¸Ń
    0.11
    è¿ijè§Ĩ
    0.10
    ä¸ĵçĶ¨
    0.10
    çľ¼æ³ª
    0.10
    æĪ´ä¸Ĭ
    0.10
    ç«ĭä½ĵ
    0.10
    æĤ£
    0.09
    严éĩįçļĦ
    0.09
    Activations Density 0.075%

    No Known Activations