© 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. 57492
    Prev
    Next
    INDEX
    Explanations

    [something]-based [noun]

    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
    èĮ¹
    -0.09
    _sid
    -0.09
    åľ¨æĦı
    -0.09
    tro
    -0.08
     Yus
    -0.08
    utations
    -0.08
     cara
    -0.08
    aire
    -0.08
    ansion
    -0.08
    jp
    -0.08
    POSITIVE LOGITS
    çİĦ
    0.10
    ä»Ĺ
    0.09
     Datum
    0.08
    hton
    0.08
    忽
    0.08
    ogne
    0.08
    áĥ£
    0.08
    authority
    0.08
    ä¸Ī
    0.08
    illion
    0.08
    Activations Density 0.109%

    No Known Activations