© 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. 4862
    Prev
    Next
    INDEX
    Explanations

    media outlets and organizations

    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
    ermann
    -0.09
    enson
    -0.09
    åıĮçľ¼
    -0.08
    zelf
    -0.08
    bine
    -0.08
    èĥļ
    -0.08
    ì¦Ī
    -0.08
    Ŀ
    -0.07
    sWith
    -0.07
     Lantern
    -0.07
    POSITIVE LOGITS
    æķ´é¡¿
    0.11
    ical
    0.10
    lymp
    0.09
    orsche
    0.09
    ç¼Ĺ
    0.09
    gne
    0.09
    centage
    0.09
    itional
    0.09
    ROTO
    0.09
    Soft
    0.09
    Activations Density 0.006%

    No Known Activations