© 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. 29854
    Prev
    Next
    INDEX
    Explanations

    why [verb] [noun]

    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
    éĿŀ常
    -0.09
    åľĪ
    -0.09
    æĶ¶åĽŀ
    -0.09
    åıĮéĩį
    -0.09
    ä¹Łæĺ¯éĿŀ常
    -0.08
    ä»İåݻ年
    -0.08
    ç²¥
    -0.08
    éĿŀ常åĸľæ¬¢
    -0.08
    orted
    -0.08
    æŨ
    -0.08
    POSITIVE LOGITS
    ä¼łæİĪ
    0.10
    ä¼ļ产çĶŁ
    0.10
    pty
    0.10
     Greeks
    0.10
    åģıåģı
    0.09
    AGENT
    0.09
    enza
    0.09
    éĢıæ°Ķ
    0.09
    èĥ½æ»¡è¶³
    0.09
    åŃĶéĽĢ
    0.09
    Activations Density 0.030%

    No Known Activations