© 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. 62027
    Prev
    Next
    INDEX
    Explanations

    for the better

    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
    rone
    -0.10
    ä¹IJæĦı
    -0.09
    翻身
    -0.09
    åīįåIJİ
    -0.09
    ç¥Ŀ
    -0.09
    ä¹IJ
    -0.09
    æĨİ
    -0.08
    prot
    -0.08
    friendly
    -0.08
    Friendly
    -0.08
    POSITIVE LOGITS
    .dtd
    0.10
    åİŁåĽłä¹ĭä¸Ģ
    0.10
    ä¸Ģ缴éĥ½æĺ¯
    0.09
    æµĶ
    0.09
    THON
    0.09
    æ¡£
    0.09
    æ°¸ä¹ħ
    0.09
    åħ¬åı¸å°Ĩ
    0.09
    åĺ´éĩĮ
    0.09
    æ§ļ
    0.09
    Activations Density 0.008%

    No Known Activations