© 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. 55979
    Prev
    Next
    INDEX
    Explanations

    learn about something new

    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
    ean
    -0.10
    æij¸
    -0.10
    èĢħ
    -0.09
    èľĢ
    -0.09
    æĶĢçĻ»
    -0.09
    å¼ķ导
    -0.08
    -touch
    -0.08
    esa
    -0.08
    áŀĹ
    -0.08
    amp
    -0.08
    POSITIVE LOGITS
    æľīä»·åĢ¼
    0.11
    æľīèī²
    0.11
     valuable
    0.10
    æľīä»·åĢ¼çļĦ
    0.10
    äºĨä¸įå°ij
    0.10
    æĦıæĥ³ä¸įåĪ°
    0.10
    äºĨå¾Īå¤ļ
    0.09
    ç¥
    0.09
    èī¯
    0.09
    棺
    0.09
    Activations Density 0.035%

    No Known Activations