© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20981
    Prev
    Next
    INDEX
    Explanations

    said object

    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
    ä»Ļ女
    -0.10
    åľ°å¸¦
    -0.09
    å°¤åħ¶æĺ¯
    -0.09
    åĨĻä¸ĭ
    -0.08
    çľ¨çľ¼
    -0.08
    éĻIJ度
    -0.08
    (branch
    -0.08
    纪念ç¢ij
    -0.08
    æĹĹä¸ĭ
    -0.08
    tern
    -0.08
    POSITIVE LOGITS
    zelf
    0.15
    tempts
    0.10
    ä¸Ģè´¯
    0.10
    лен
    0.09
    /licenses
    0.09
    лок
    0.09
    æĸ°é¢ĸ
    0.09
    à¹ģà¸ŀ
    0.09
    äºĭçī©
    0.09
    èĩ¾
    0.09
    Activations Density 0.018%

    No Known Activations