© 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. 49102
    Prev
    Next
    INDEX
    Explanations

    book compounds

    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
    è°¤
    -0.10
    ละ
    -0.09
    è£ı
    -0.08
    пÑĥ
    -0.08
    å¼ĺæī¬
    -0.08
    CF
    -0.08
    æľīä¿¡å¿ĥ
    -0.08
     Gda
    -0.08
    Morning
    -0.08
    ä¸ĵä¸ļæĬĢæľ¯
    -0.08
    POSITIVE LOGITS
    rey
    0.10
    ings
    0.10
    athon
    0.09
    ä¸Ģä¸Ŀ
    0.09
    sl
    0.09
    _soup
    0.09
    äºĨä¸Ģåľº
    0.09
    ä¸Ģåľº
    0.09
    ä¸į好
    0.09
    ERY
    0.08
    Activations Density 0.014%

    No Known Activations