© 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. 11590
    Prev
    Next
    INDEX
    Explanations

    assumed office, responsibility, or role

    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
    wire
    -0.10
    æ´»å¾Ĺ
    -0.08
    iculty
    -0.08
    lund
    -0.08
    ening
    -0.07
    ness
    -0.07
    ÙĨÙī
    -0.07
    hiba
    -0.07
    otten
    -0.07
    宽æķŀ
    -0.07
    POSITIVE LOGITS
     responsibility
    0.12
    社ä¼ļ责任
    0.11
    æ¼Ĥæµģ
    0.11
    责任
    0.11
    æ¨Ĺ
    0.10
    åºĶæľīçļĦ
    0.10
    æī¹è¯Ħ
    0.09
    å¤ļç§įå½¢å¼ı
    0.09
     Responsibility
    0.09
    ä¹Ŀé¾Ļ
    0.09
    Activations Density 0.014%

    No Known Activations