© 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. 21137
    Prev
    Next
    INDEX
    Explanations

    hide, conceal, cover, disguise

    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
    .ssl
    -0.09
     ÑģÑĤо
    -0.09
     bbw
    -0.09
    å¥ī
    -0.09
    Defense
    -0.08
    -TV
    -0.08
    Wildcard
    -0.08
    æİĮæı¡
    -0.08
    ISCO
    -0.08
    Jacob
    -0.08
    POSITIVE LOGITS
    å¿Ļ
    0.11
    羣缸
    0.11
    åĮĨ
    0.10
    亮çĤ¹
    0.10
    .Hide
    0.10
    ä½ıäºĨ
    0.09
    è¿ĺ以为
    0.09
    æݩ饰
    0.09
    åĭĥåĭĥ
    0.09
    ä¸ij
    0.09
    Activations Density 0.067%

    No Known Activations