© 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. 57766
    Prev
    Next
    INDEX
    Explanations

    ages and time periods

    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
    è¿ĩ大
    -0.12
    cry
    -0.09
    åĨ¤
    -0.09
    åĪ»æĦı
    -0.09
    æĺĮ
    -0.09
    odon
    -0.09
    éĴ¡
    -0.09
    quate
    -0.09
    çĪ¹
    -0.08
    åѤ
    -0.08
    POSITIVE LOGITS
    éĴī
    0.11
    ä¸ĢäºĮ
    0.11
    è°¤
    0.09
    年纪
    0.09
    å·¥ä½ľç«Ļ
    0.08
    ä¸īåĽĽ
    0.08
    ÙģØŃ
    0.08
    isObject
    0.08
    æĪĸå¤ļæĪĸå°ij
    0.08
    åĩłåIJį
    0.08
    Activations Density 0.042%

    No Known Activations