© 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. 18835
    Prev
    Next
    INDEX
    Explanations

    papers, abstract, colons

    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
    æIJ
    -0.09
    人类
    -0.09
    è¿Ļç¯ĩ
    -0.09
    inox
    -0.09
     Mist
    -0.09
    isObject
    -0.09
    specs
    -0.09
    大åŁİå¸Ĥ
    -0.09
    ROY
    -0.09
     Ferd
    -0.08
    POSITIVE LOGITS
    æľīä¸įå°ij
    0.10
    aily
    0.09
    笥
    0.09
    为主线
    0.09
    æľīä¸Ģ
    0.09
    frog
    0.09
    ä¼
    0.09
     Garn
    0.09
    hower
    0.09
    éĤ£ä¸Ģ
    0.08
    Activations Density 0.038%

    No Known Activations