© 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. 11257
    Prev
    Next
    INDEX
    Explanations

    wit, witty, Witt, witticism

    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
    ร
    -0.11
    ever
    -0.10
    aur
    -0.09
    çĶļèĩ³è¿ŀ
    -0.09
    å¿ĥ头
    -0.09
    OrUpdate
    -0.09
    mit
    -0.09
    讨
    -0.08
    -den
    -0.08
    ä¹IJéĺŁ
    -0.08
    POSITIVE LOGITS
    fully
    0.12
    ../../../
    0.10
    marsh
    0.10
    ively
    0.10
    folio
    0.10
    iesz
    0.09
    ful
    0.09
    æ¹ĸåĮº
    0.09
    rell
    0.09
    fulness
    0.09
    Activations Density 0.004%

    No Known Activations