© 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. 10798
    Prev
    Next
    INDEX
    Explanations

    heresy, blasphemous, sacrilegious

    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
    èİ«åIJįåħ¶å¦Ļ
    -0.10
    夫人
    -0.10
    èĢģ人家
    -0.09
    èİ«åIJį
    -0.09
    究
    -0.09
    è¦ĨçĽĸçİĩ
    -0.09
    ä¸įèµ·
    -0.09
    è´ŀ
    -0.09
    æĪĸå¤ļ
    -0.08
    彬
    -0.08
    POSITIVE LOGITS
    èĢħçļĦ
    0.12
    èĢħ
    0.11
    ously
    0.09
    æİī
    0.09
    éŃģ
    0.09
    ãģªãĤĬ
    0.08
    ศา
    0.08
    éĽĨåĽ¢èĤ¡ä»½
    0.08
    ively
    0.08
    æİīäºĨ
    0.08
    Activations Density 0.010%

    No Known Activations