© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 29544
    Prev
    Next
    INDEX
    Explanations

    kirk, kirke, or ken names

    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
    Negative Logits
    ÑģÑĮ
    -0.13
    çĸĻ
    -0.11
    éĶľ
    -0.10
     пÑĢоÑģмоÑĤÑĢ
    -0.10
    家ä¼Ļ
    -0.10
    ÑĤÑĮ
    -0.10
    ä¸įäºĨ
    -0.09
    --------------------------------------------------------------------------↵
    -0.09
    ika
    -0.09
    éĿ¢ä¸Ĭ
    -0.09
    POSITIVE LOGITS
    ness
    0.12
    patrick
    0.11
    side
    0.11
    shire
    0.10
    åĬŀåħ¬åİħ
    0.09
    sville
    0.09
    jab
    0.09
    ains
    0.09
    ол
    0.09
    涯
    0.09
    Activations Density 0.012%

    No Known Activations