© 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. 10648
    Prev
    Next
    INDEX
    Explanations

    bring together

    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¿Ľåħ¥äºĨ
    -0.11
    onal
    -0.10
    ç«ł
    -0.10
    巴士
    -0.09
    ارة
    -0.09
    taboola
    -0.09
    åı¸
    -0.09
    евиÑĩ
    -0.09
    .eye
    -0.09
    æij§
    -0.09
    POSITIVE LOGITS
    èģļéĽĨ
    0.11
     together
    0.10
    åľ¨ä¸Ģèµ·
    0.10
    ped
    0.09
    å¼ĢæĿ¥
    0.09
    hoa
    0.09
    éĽĨèģļ
    0.08
     Ry
    0.08
    æ±ĩéĽĨ
    0.08
    æ±ĩèģļ
    0.08
    Activations Density 0.036%

    No Known Activations