© 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. 63300
    Prev
    Next
    INDEX
    Explanations

    lethal threat dangerous feared

    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¯ĨéĽĨ
    -0.09
    à¸ģำ
    -0.09
    轨
    -0.09
    佬
    -0.08
     tide
    -0.08
    çĸı
    -0.08
    åĨľåľº
    -0.08
     strokeLine
    -0.08
    ug
    -0.08
    ç¼ĺåĪĨ
    -0.08
    POSITIVE LOGITS
    åįģ足
    0.10
    ä¸įæİĴéĻ¤
    0.09
    ness
    0.09
    éĿ
    0.09
    inc
    0.09
    ìĿĮìĿĦ
    0.08
    宥
    0.08
    æīĪ
    0.08
    aine
    0.08
     Invocation
    0.08
    Activations Density 0.013%

    No Known Activations