© 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. 20301
    Prev
    Next
    INDEX
    Explanations

    pancakes

    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
    geist
    -0.11
    iday
    -0.10
    åıµ
    -0.09
    perature
    -0.09
    å»ĵ
    -0.09
    legate
    -0.09
    è²Į
    -0.09
    ãģĹãģ¾ãģĨ
    -0.09
    æ¢Ĺ
    -0.09
    ials
    -0.08
    POSITIVE LOGITS
    uts
    0.10
    ä¸ĢåĽŀäºĭ
    0.10
    iates
    0.09
    åĽŀäºĭ
    0.08
    ettes
    0.08
    éĽ¹
    0.08
    çħĮ
    0.08
    uth
    0.08
    ormal
    0.08
    ä¸įä¸Ĭ
    0.08
    Activations Density 0.007%

    No Known Activations