© 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. 20297
    Prev
    Next
    INDEX
    Explanations

    appetizers or appetizing

    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
    ä¸Ĭä¸ĭ游
    -0.10
    éŀį
    -0.10
     autorelease
    -0.09
    auss
    -0.09
    noch
    -0.09
    utch
    -0.09
    éªĸ
    -0.09
     inters
    -0.09
    EEE
    -0.09
    aten
    -0.09
    POSITIVE LOGITS
    æĪijçĽ¸ä¿¡
    0.10
    rete
    0.09
    åħĪè¡Į
    0.09
    棣
    0.09
    Wars
    0.09
    füg
    0.08
    è¿ĻåĿĹ
    0.08
    åĽ½åĬ¡
    0.08
    模èĮĥ
    0.08
    zon
    0.08
    Activations Density 0.006%

    No Known Activations