© 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. 31841
    Prev
    Next
    INDEX
    Explanations

    smoothie or shake

    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
    urat
    -0.11
    麻å°Ĩ
    -0.10
    stell
    -0.09
     Strom
    -0.09
    éķij
    -0.09
    GTK
    -0.09
    én
    -0.09
     xấu
    -0.09
    åįķè¯į
    -0.08
    毫ä¸į
    -0.08
    POSITIVE LOGITS
    æ¸į
    0.08
    ä¸ĢæĹı
    0.08
    FER
    0.08
    fusion
    0.08
    ifter
    0.08
    çļĦç§ĺå¯Ĩ
    0.08
    çİĭåºľ
    0.08
    èĽ°
    0.08
    ä¸įä¸Ĭ
    0.08
    æºŁ
    0.07
    Activations Density 0.004%

    No Known Activations