© 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. 35048
    Prev
    Next
    INDEX
    Explanations

    whole group, life, career, day, money

    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
    èĢģèĻİ
    -0.08
     whole
    -0.08
    ese
    -0.08
    ion
    -0.08
     вполне
    -0.08
    ç§įç§į
    -0.08
    å¤ļæĸ¹
    -0.08
    asan
    -0.08
    æĽ´å¤ļçļĦ
    -0.08
    ç©°
    -0.08
    POSITIVE LOGITS
    maal
    0.13
    uded
    0.12
    /fast
    0.11
    代表大ä¼ļ
    0.10
    endet
    0.10
    æŶéĹ´åĨħ
    0.10
    åıĪ好
    0.10
    æĦıä¹īä¸ĬçļĦ
    0.10
    æĦıä¹īä¸Ĭ
    0.10
    sworth
    0.09
    Activations Density 0.017%

    No Known Activations