© 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. 1637
    Prev
    Next
    INDEX
    Explanations

    medieval history

    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
    岫
    -0.08
     cork
    -0.08
    aho
    -0.08
    .uf
    -0.08
    串
    -0.08
    linky
    -0.08
    ahoo
    -0.08
    é¬ĥ
    -0.08
     Dayton
    -0.08
     Venezuel
    -0.08
    POSITIVE LOGITS
    为空
    0.10
    èĭ±æł¼åħ°
    0.10
    æįº
    0.10
    æīĪ
    0.09
    紧缺
    0.09
    é²±
    0.09
    ç»įåħ´
    0.09
     Medieval
    0.09
    æĪĸå¤ļ
    0.09
    é¹ĺ
    0.09
    Activations Density 0.181%

    No Known Activations