© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19714
    Prev
    Next
    INDEX
    Explanations

    chameleon, tango, juggle, act, ride

    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
    è²Į
    -0.11
    nings
    -0.10
    Comparer
    -0.09
    stro
    -0.09
    æĬķåħ¥åĪ°
    -0.08
    æĺĨä»ij
    -0.08
    èĢĢ
    -0.08
    ileged
    -0.08
    uci
    -0.08
    paces
    -0.08
    POSITIVE LOGITS
    ç©¿æ¢Ń
    0.10
    èĽĻ
    0.10
    ä¹¾åĿ¤
    0.09
    è··
    0.09
    å¼ıçļĦ
    0.09
    èĪ¬çļĦ
    0.09
    baz
    0.09
    ä¸ĢåĪĻ
    0.09
    ä¸Ģç¯ĩæĸĩç«ł
    0.08
    åīĶ
    0.08
    Activations Density 0.035%

    No Known Activations