© 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. 19310
    Prev
    Next
    INDEX
    Explanations

    painting, long, see, color

    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
    å¿«æį·
    -0.10
    è®°ä½ı
    -0.10
    æĹŃ
    -0.10
    ç¬ijçĿĢ说
    -0.10
    .Configure
    -0.09
    éĵ
    -0.09
    âĸĬ
    -0.09
     vö
    -0.09
    æŁĦ
    -0.09
    ä»°
    -0.09
    POSITIVE LOGITS
    ä¹ŁæŃ£æĺ¯
    0.08
    æľ¬èº«å°±
    0.08
    æīį
    0.08
    ajo
    0.08
    æīĢ以
    0.07
    ADER
    0.07
     naturally
    0.07
     But
    0.07
     Fig
    0.07
    uby
    0.07
    Activations Density 1.746%

    No Known Activations