© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37751
    Prev
    Next
    INDEX
    Explanations

    killing people or murder

    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
    å´İ
    -0.09
    SCAN
    -0.09
    scan
    -0.09
    sw
    -0.09
    è¾ŀ
    -0.08
    èĥĮ
    -0.08
    éĿ¢åIJij
    -0.08
    è¾Ń
    -0.08
    .scan
    -0.08
    çIJĽ
    -0.08
    POSITIVE LOGITS
    éĶı
    0.12
    åĿ¯
    0.11
    åĢĴåľ¨
    0.11
    æĿĢäºĨ
    0.10
    æĪ®
    0.10
    liest
    0.10
    æİī
    0.10
    æĿĢ
    0.09
    çģŃ
    0.09
    ä½łçľĭ
    0.09
    Activations Density 0.092%

    No Known Activations