© 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. 5186
    Prev
    Next
    INDEX
    Explanations

    break, broke, breaks

    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
    åİĨå¹´
    -0.09
    /of
    -0.08
    ä¸įåľ¨
    -0.08
    ä½ľèĢħæīĢæľī
    -0.08
    sj
    -0.08
     Gems
    -0.08
    iage
    -0.08
    izz
    -0.08
    gte
    -0.08
     craz
    -0.08
    POSITIVE LOGITS
    éĭĨ
    0.10
    éĩįæķ´
    0.10
    å¾Ĺå¾Ī
    0.09
    fuse
    0.09
    eree
    0.09
    rive
    0.08
    窿
    0.08
    zeich
    0.08
    á»Ī
    0.08
    çļĦ羣å®ŀ
    0.08
    Activations Density 0.019%

    No Known Activations