© 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. 8584
    Prev
    Next
    INDEX
    Explanations

    struck me

    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
    trs
    -0.10
     вÑģÑĤÑĢ
    -0.09
    ãģĵãģĿ
    -0.09
    å·§
    -0.08
    èĬ¯
    -0.08
     TSR
    -0.08
    ä¸Ńæľī
    -0.08
    nst
    -0.08
    ä¸įä¸Ģå®ļ
    -0.08
    æ±
    -0.08
    POSITIVE LOGITS
    å¼Ĥ
    0.10
    讶
    0.10
    çķ°
    0.09
    æĦŁåı¹
    0.09
    çļĦåį°è±¡
    0.09
    é¢ĵ
    0.09
     struck
    0.09
    åĬŀ好
    0.09
    æģ¯
    0.09
    atti
    0.09
    Activations Density 0.018%

    No Known Activations