© 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. 23327
    Prev
    Next
    INDEX
    Explanations

    try one

    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
    mx
    -0.09
    arking
    -0.08
    jab
    -0.08
    ople
    -0.08
    éĢĹ
    -0.08
    aset
    -0.08
    edly
    -0.08
    åĸĩ
    -0.07
    éĿĴå±±
    -0.07
    åŃ
    -0.07
    POSITIVE LOGITS
    åIJĪãĤı
    0.09
    çļĦæľºä¼ļ
    0.09
    AIR
    0.09
    çIJĨ财产åĵģ
    0.09
    çļĦä¸ľè¥¿
    0.09
    æīŀ
    0.09
    å¹Ķ
    0.09
    -toggler
    0.09
    åıī
    0.09
    (Stack
    0.08
    Activations Density 0.030%

    No Known Activations