© 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. 1487
    Prev
    Next
    INDEX
    Explanations

    code assignments with punctuation

    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
     starters
    -0.08
    ISH
    -0.08
     pot
    -0.07
    群
    -0.07
    è¶Ĭæĺ¯
    -0.07
    iers
    -0.07
    .just
    -0.06
    ä¸Ńå°ıåŀĭ
    -0.06
    place
    -0.06
    ãĢģ
    -0.06
    POSITIVE LOGITS
    ä¾Ħ
    0.10
    æİĮæŁľ
    0.09
    红éħĴ
    0.08
    è¿Ļæł·ä¸ĢæĿ¥
    0.08
    èķ¨
    0.08
    éģĤ
    0.08
    æĪ¿éĹ´éĩĮ
    0.08
    riteln
    0.08
    elivery
    0.08
    ovy
    0.08
    Activations Density 1.305%

    No Known Activations