© 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. 48216
    Prev
    Next
    INDEX
    Explanations

    less, not

    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
    俨
    -0.10
    ä¾ĽåĽ¾
    -0.09
    大æ¦Ĥ
    -0.08
    大å®Ĺ
    -0.08
    rons
    -0.08
    _TERM
    -0.08
    åIJµ
    -0.08
    æľ¬æĬ¥
    -0.08
    æŀľçĦ¶
    -0.08
    å§ij
    -0.08
    POSITIVE LOGITS
    kah
    0.09
    æĺ¯åľ¨
    0.09
    åºĬä¸Ĭ
    0.09
    æĮ£æīİ
    0.09
    geh
    0.09
     Gim
    0.09
    èĩ³æĸ¼
    0.08
    çĸ³
    0.08
    åIJİéĿ¢çļĦ
    0.08
     ones
    0.08
    Activations Density 0.010%

    No Known Activations