© 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. 57192
    Prev
    Next
    INDEX
    Explanations

    names with photo credits

    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
    lte
    -0.08
     Gore
    -0.08
    åıįä¹ĭ
    -0.08
     manh
    -0.08
     po
    -0.08
    FormattedMessage
    -0.08
     Odyssey
    -0.07
    ded
    -0.07
    丰满
    -0.07
    ppo
    -0.07
    POSITIVE LOGITS
    åĽ¾çīĩ
    0.10
    æĬķèµĦåŁºéĩij
    0.10
    æĭ¿äºĨ
    0.10
    åľĸçīĩ
    0.09
    沫
    0.09
    ç»´å¥ĩ
    0.09
    nger
    0.09
    ancel
    0.09
    æĭ¿çĿĢ
    0.09
    ä¾µçĬ¯
    0.09
    Activations Density 0.031%

    No Known Activations