© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9681
    Prev
    Next
    INDEX
    Explanations

    reproduce replication reproducibility

    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
    å¼ĥ
    -0.11
    anie
    -0.10
    shot
    -0.09
    ç²¾ç¥ŀæĸĩæĺİ
    -0.09
    деж
    -0.09
    .FileSystem
    -0.09
    eland
    -0.09
    otte
    -0.09
    ей
    -0.09
    .ali
    -0.09
    POSITIVE LOGITS
    模æĭŁ
    0.16
    éĩįçİ°
    0.15
    åĨįçİ°
    0.14
    çݯå¢ĥä¸ĭ
    0.13
     conditions
    0.12
    æķħæĦı
    0.12
    æĿ¡ä»¶ä¸ĭ
    0.12
    模仿
    0.11
    çݯå¢ĥ
    0.11
     reproduce
    0.11
    Activations Density 0.290%

    No Known Activations