© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50788
    Prev
    Next
    INDEX
    Explanations

    original, your, other, anti, unwilling, real, ultimate, 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
    æĴĴ
    -0.10
    æīĵéĢļ
    -0.09
    è¶Ĭ
    -0.08
    ç¼ł
    -0.08
    dek
    -0.08
    纯
    -0.08
    ÙħØ®
    -0.08
     nic
    -0.08
    æĬij
    -0.08
    åĩ¶
    -0.08
    POSITIVE LOGITS
    èĭ±éĽĦ
    0.12
    æĦıä¹īä¸ĬçļĦ
    0.11
    å®ļä½į
    0.10
     hero
    0.09
    æľĢæĹ©çļĦ
    0.09
    è¯ł
    0.09
    èıģ
    0.09
    ä¸Ńä»ĭ
    0.09
    é¢ijé¢ij
    0.09
    æľŁåĨħ
    0.09
    Activations Density 0.157%

    No Known Activations