© 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. 48416
    Prev
    Next
    INDEX
    Explanations

    .rest or .space domains

    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
    å°ĨæĿ¥
    -0.10
    大家éĥ½
    -0.10
    tent
    -0.09
    éĻIJ
    -0.08
    æ³¥åľŁ
    -0.08
    大家ä¸Ģèµ·
    -0.08
    éļ¾å¾Ĺ
    -0.08
    æ³¥
    -0.08
    mach
    -0.08
    夫å¦ĩ
    -0.08
    POSITIVE LOGITS
     Outcome
    0.10
     Cly
    0.10
    ystore
    0.09
    ä¸ļæĢģ
    0.09
    Outcome
    0.09
    èĻ±
    0.09
    .sd
    0.09
    ospace
    0.08
     Coco
    0.08
    å¾ī
    0.08
    Activations Density 0.001%

    No Known Activations