© 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
    2. Qwen3-32B
    3. 32-RESID-BATCHTOPK-65K
    4. 38605
    Prev
    Next
    INDEX
    Explanations

    pretty [adjective]

    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
    æķħæĦı
    -0.10
    è¿Ļæīį
    -0.09
    izo
    -0.09
     ìĺ¤íŀĪ볤
    -0.09
    éĹªç͵
    -0.09
    è¢ħ
    -0.09
    æĽ´å¤ļä¿¡æģ¯
    -0.09
    è¿Ļä¸Ģ天
    -0.09
    бал
    -0.09
    IMAL
    -0.09
    POSITIVE LOGITS
    ä¸įéĶĻçļĦ
    0.15
    ly
    0.11
    大çļĦ
    0.10
    æ·±åĪ»çļĦ
    0.10
    åı¯è§Ĥ
    0.10
    ä¸įå°ıçļĦ
    0.10
    oting
    0.10
    好çļĦ
    0.09
    ä¸įéĶĻ
    0.09
    gene
    0.09
    Activations Density 0.013%

    No Known Activations