© 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. 33461
    Prev
    Next
    INDEX
    Explanations

    purple cat

    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
    eward
    -0.09
    绾
    -0.09
    å©Ĭ
    -0.08
    æīķãģĦ
    -0.08
    妯
    -0.08
    çĻ½é©¬
    -0.08
    鸾
    -0.08
    大çIJĨçŁ³
    -0.08
    èĨĿ
    -0.08
    ä¸Ĭåīį
    -0.08
    POSITIVE LOGITS
    æĹı
    0.11
    大象
    0.09
    opol
    0.09
    伸
    0.09
    个æĢ§
    0.08
    åĿ¦
    0.08
    æĨ¨
    0.08
    弯
    0.08
    shape
    0.08
    æĹłåı¯
    0.08
    Activations Density 0.123%

    No Known Activations