© 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. 57243
    Prev
    Next
    INDEX
    Explanations

    is a

    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
     themselves
    -0.11
    éĥ½å¾Ī
    -0.10
    å¥Ħ
    -0.10
    éĥ½éĿŀ常
    -0.10
    ä¹ĥ
    -0.09
    å®ĥåı¯ä»¥
    -0.09
    æĹłåı¯
    -0.09
    京津
    -0.09
     иÑħ
    -0.09
    ä¹ĭå¿ĥ
    -0.08
    POSITIVE LOGITS
    èĢģ人家
    0.14
    iner
    0.11
     incumbent
    0.11
    inerary
    0.11
     raining
    0.10
    æĪIJ为ä¸Ģ个
    0.10
    /th
    0.10
    ä½ľä¸ºä¸Ģç§į
    0.10
    chy
    0.09
    he
    0.09
    Activations Density 1.541%

    No Known Activations