© 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. 23037
    Prev
    Next
    INDEX
    Explanations

    was, ar, i was, i as

    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
    åĨł
    -0.09
    æĬĬè¿Ļ个
    -0.09
    Operand
    -0.09
     inde
    -0.08
    é¢į
    -0.08
    ç¨Ĺ
    -0.08
    ovan
    -0.08
     van
    -0.08
    à¹Ĩ
    -0.08
    _errno
    -0.08
    POSITIVE LOGITS
    ALER
    0.11
    ä¾ĭå¤ĸ
    0.10
    /is
    0.10
    izes
    0.09
    ras
    0.09
    åĴĢ
    0.09
    äºĭåIJİ
    0.09
    -UA
    0.09
    odia
    0.08
    à¸Īำ
    0.08
    Activations Density 0.109%

    No Known Activations