© 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. 39081
    Prev
    Next
    INDEX
    Explanations

    life-threatening

    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
    è´Ńç½®
    -0.10
    Soup
    -0.09
    é¥Ń
    -0.09
    æĵ¦æĭŃ
    -0.09
    èĩªçĦ¶èĢĮ
    -0.08
    ãģ¤ãģĦ
    -0.08
    æĶ¾å¼Ģ
    -0.08
    åĴĮæĬĢæľ¯
    -0.08
    æ°Ķåij³
    -0.08
    综
    -0.08
    POSITIVE LOGITS
    åį±
    0.13
    å¨ģèĥģ
    0.12
    æĢ¥éľĢ
    0.11
     threatening
    0.11
    -threatening
    0.11
     urgent
    0.10
    åįģ大
    0.10
    eneg
    0.09
    çĶŁæŃ»
    0.09
    -threat
    0.09
    Activations Density 0.117%

    No Known Activations