© 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. 36130
    Prev
    Next
    INDEX
    Explanations

    money laundering

    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
    _ANY
    -0.10
    飶
    -0.09
    èĦ¾
    -0.09
    åı®
    -0.09
    ellan
    -0.09
    屯
    -0.09
    æĢĢ
    -0.09
    tems
    -0.08
    _rng
    -0.08
    ä¾ĽåĽ¾
    -0.08
    POSITIVE LOGITS
    æݩ饰
    0.12
    æµģåħ¥
    0.11
    æŃ£å¸¸ä½¿çĶ¨
    0.10
    奢
    0.10
     Lau
    0.10
    å®ŀéĻħæݧåĪ¶
    0.09
    flow
    0.09
     flow
    0.09
    交å¾Ģ
    0.09
    å¾ĢæĿ¥
    0.09
    Activations Density 0.034%

    No Known Activations