© 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. 8269
    Prev
    Next
    INDEX
    Explanations

    current status or situation

    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
    åĿļå®Ī
    -0.10
    æIJŀ
    -0.09
    dest
    -0.08
    æĭĨ
    -0.08
    à¥į
    -0.08
    -origin
    -0.08
    clo
    -0.08
    rest
    -0.08
    [assembly
    -0.08
     Dest
    -0.08
    POSITIVE LOGITS
    "),"
    0.10
    æĪĸå°Ĩ
    0.09
    ä¸įæĪIJ
    0.09
     اÙĦÙĨÙĤ
    0.09
    éĺĴ
    0.09
    åĨħ马
    0.09
     ÙĪÙĦÙħ
    0.09
    ä¸į代表
    0.09
    æĥ¯
    0.09
    æı©
    0.09
    Activations Density 0.003%

    No Known Activations