© 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. 36908
    Prev
    Next
    INDEX
    Explanations

    Wigan Athletic, Mirror's Edge Catalyst, MediaPipe Authors

    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
    äºī龸
    -0.10
     Lá»ĭch
    -0.09
    ä¼ī
    -0.09
    éłĺ
    -0.09
    _nb
    -0.09
    camp
    -0.09
    æīĵæŀ¶
    -0.08
     Warner
    -0.08
    æľ´ç´ł
    -0.08
    aby
    -0.08
    POSITIVE LOGITS
    çŃ¹èµĦ
    0.09
    uilt
    0.09
    æł¸å¿ĥæĬĢæľ¯
    0.09
    èķĥ
    0.09
    è¿Ķ
    0.09
    DEFINE
    0.09
    osa
    0.09
    å½ĵä¸ĭ
    0.09
    efined
    0.08
    amac
    0.08
    Activations Density 0.033%

    No Known Activations