© 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. 34965
    Prev
    Next
    INDEX
    Explanations

    severe enoughlarge enoughstrong enoughpowerful enoughhigh enoughhigh abovenot enoughnot enoughlong enoughadvanced enough

    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
    åľ¨å®¶éĩĮ
    -0.10
    å¹´åºķåīį
    -0.09
    æľªæĿ¥çļĦ
    -0.09
    çİ°åľ¨çļĦ
    -0.09
    YS
    -0.09
    人æ°ijæľįåĬ¡
    -0.09
    è¾¾å°Ķ
    -0.09
    ä¸¥æł¼
    -0.09
    _fil
    -0.08
    æīĢåľ¨çļĦ
    -0.08
    POSITIVE LOGITS
     enough
    0.31
    è¶³å¤Ł
    0.25
    å¤Ł
    0.22
     Enough
    0.22
    è¶³å¤ŁçļĦ
    0.20
    Enough
    0.20
    å¤ł
    0.17
    ä¸įå¤Ł
    0.16
    足以
    0.15
     Äijủ
    0.14
    Activations Density 0.041%

    No Known Activations