© 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. 11885
    Prev
    Next
    INDEX
    Explanations

    describing or characterizing qualities

    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
    ATER
    -0.09
    Minor
    -0.08
     Minor
    -0.08
    æ³Ľ
    -0.08
    ncia
    -0.08
    ä¹łæĥ¯
    -0.08
     Lone
    -0.08
    é»ĦçŁ³
    -0.08
    æ¦Ĩ
    -0.08
    èIJ½
    -0.08
    POSITIVE LOGITS
    形容
    0.17
    æĿ¥å½¢å®¹
    0.17
    åħ³éĶ®è¯į
    0.15
    çī¹çĤ¹æĺ¯
    0.13
    çī¹è´¨
    0.12
    è¯įè¯Ń
    0.12
    æııè¿°
    0.11
    ä¸Ģåı¥è¯Ŀ
    0.11
    çī¹çĤ¹
    0.11
     ÑħаÑĢакÑĤеÑĢиÑģÑĤик
    0.10
    Activations Density 0.050%

    No Known Activations