© 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. 32608
    Prev
    Next
    INDEX
    Explanations

    forum threads

    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
    大æĸ¹
    -0.10
    æĤį
    -0.09
     HO
    -0.08
    loo
    -0.08
     Camp
    -0.08
    åķ¼
    -0.08
    éĺŁ
    -0.08
    ãģ¨ãģĦãģ£ãģ¦
    -0.08
    licity
    -0.08
    Tho
    -0.08
    POSITIVE LOGITS
    [thread
    0.12
    oola
    0.11
    å¼ı
    0.10
    zelf
    0.10
    æĪĸå¤ļ
    0.09
    .currentThread
    0.09
    åŁ
    0.09
     THREAD
    0.09
    à´³
    0.09
    kon
    0.09
    Activations Density 0.019%

    No Known Activations