© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52105
    Prev
    Next
    INDEX
    Explanations

    encouraged, welcome, allowed, responsible, requested

    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
    éĻªä½ł
    -0.10
    åīįçŀ»
    -0.10
    æĪijèĥ½
    -0.09
    è¡Ķ
    -0.09
    ä½łæĥ³
    -0.09
    ä¸įå®Į
    -0.09
    æľīäºĽäºº
    -0.09
    æ·±åħ¥äººå¿ĥ
    -0.09
    rh
    -0.09
    好å¤Ħ
    -0.08
    POSITIVE LOGITS
    大èĥĨ
    0.10
    bench
    0.09
    imb
    0.09
    èĩªåľ¨
    0.09
    建议
    0.09
    åĮĸåѦåĵģ
    0.09
    沪
    0.09
    orate
    0.09
     hereby
    0.08
     welcome
    0.08
    Activations Density 0.042%

    No Known Activations