© 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. 58080
    Prev
    Next
    INDEX
    Explanations

    polite or politician

    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
    acer
    -0.10
    æĺŃ
    -0.10
    èħ°
    -0.09
    stä
    -0.09
    áĢļ
    -0.08
    ilton
    -0.08
    ени
    -0.08
    ses
    -0.08
    ä¸Ģè§Ī
    -0.08
    æĬ±ä½ı
    -0.08
    POSITIVE LOGITS
    äºĨä¸Ģåľº
    0.10
     ÐŃÑĤ
    0.10
    iot
    0.10
    gems
    0.09
    orno
    0.09
    æĪIJåĵģ
    0.09
    åħ¸èĮĥ
    0.09
    weets
    0.09
    AndServe
    0.09
    è´·
    0.09
    Activations Density 0.003%

    No Known Activations