© 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. 32618
    Prev
    Next
    INDEX
    Explanations

    public health and safety

    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
    éľĩ
    -0.12
     Malk
    -0.09
    é¬ĵ
    -0.09
    éŁ¿
    -0.09
    ç»ıèIJ¥çIJĨ念
    -0.09
    çģµ
    -0.08
    翻身
    -0.08
    icz
    -0.08
    幡
    -0.08
    agon
    -0.07
    POSITIVE LOGITS
    å®īåħ¨
    0.17
     safety
    0.17
     Safety
    0.15
    å®īåħ¨æĢ§
    0.15
    Safety
    0.15
    unsafe
    0.13
     unsafe
    0.13
    人身
    0.12
    Unsafe
    0.12
    à¸Ľà¸¥à¸Ńà¸Ķ
    0.12
    Activations Density 0.076%

    No Known Activations