© 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. 58244
    Prev
    Next
    INDEX
    Explanations

    offensive behavior or actions

    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
    ied
    -0.10
    òng
    -0.09
    ushima
    -0.09
    ãĤĤãģĨ
    -0.09
    çĿĽ
    -0.09
    oded
    -0.09
    foot
    -0.09
    交åıĭ
    -0.08
    ings
    -0.08
    æľĢåIJİä¸Ģ
    -0.08
    POSITIVE LOGITS
    éĥ¨éķ¿
    0.10
    леп
    0.10
    ega
    0.09
    áµı
    0.09
    -minded
    0.09
    egov
    0.09
    eny
    0.09
    eature
    0.09
    pga
    0.09
    ña
    0.08
    Activations Density 0.003%

    No Known Activations