© 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. 61603
    Prev
    Next
    INDEX
    Explanations

    characteristic, typical, normally, popular, accustomed

    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
    çĪ±ä½ł
    -0.09
    ä»İæł¹æľ¬ä¸Ĭ
    -0.08
     therefore
    -0.08
    æĪĸèĢħ说
    -0.08
     Vor
    -0.08
     MILF
    -0.08
    holm
    -0.08
     ×IJ×Ĺר
    -0.07
    ìł¤
    -0.07
     tha
    -0.07
    POSITIVE LOGITS
    æ¯ıéĢ¢
    0.10
    éĤ£ä¹Īå¤ļ
    0.10
    iko
    0.09
    éĺĤ
    0.09
    éĶ¤
    0.09
    å¹¿æ³ĽåºĶçĶ¨
    0.09
    åIJĦ级
    0.09
    åĮħåĽ´
    0.09
    æł¡å¤ĸ
    0.08
    缼è¡Į
    0.08
    Activations Density 0.174%

    No Known Activations