© 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. 51157
    Prev
    Next
    INDEX
    Explanations

    practice and camp discussions

    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¸Ģè¾ĪåŃIJ
    -0.09
    æľī害
    -0.09
    çļĦçľ¼
    -0.09
    ç»Ī身
    -0.09
    ç¤ģ
    -0.08
    ä¹ĭå¿ĥ
    -0.08
    å¤ĩ
    -0.08
    éĽĨä½ĵç»ıæµİ
    -0.08
     dipping
    -0.08
    æĦıåij³
    -0.08
    POSITIVE LOGITS
    embr
    0.10
     Sham
    0.09
    æĻĸ
    0.09
    å¿ĥå¾Ĺ
    0.09
    assed
    0.08
    人éĢī
    0.08
    åĪĨåĮĸ
    0.08
    ìŀ¥ìĿĦ
    0.08
    бол
    0.08
    ople
    0.08
    Activations Density 0.004%

    No Known Activations