© 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. 21354
    Prev
    Next
    INDEX
    Explanations

    extreme conditions or situations

    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
    æĸ¡
    -0.09
     interests
    -0.09
    lick
    -0.08
    åIJ¯è¿ª
    -0.08
    åĮ¿åIJį
    -0.08
    entrant
    -0.08
    æı©
    -0.08
    ERP
    -0.08
    绾
    -0.08
    룸
    -0.08
    POSITIVE LOGITS
    æŀģ端
    0.33
     extremes
    0.28
     extreme
    0.25
     Extreme
    0.23
    Extreme
    0.23
     extrem
    0.19
    æŀģèĩ´
    0.19
    å¤¸å¼ł
    0.18
     кÑĢай
    0.18
     extremism
    0.16
    Activations Density 0.075%

    No Known Activations