© 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. 1070
    Prev
    Next
    INDEX
    Explanations

    Ibn, al-, Iran, tribe, Jews

    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
    pad
    -0.10
    ä¸įéĶĻ
    -0.09
    好å¥ĩå¿ĥ
    -0.09
    _PAD
    -0.09
    ç²ī
    -0.09
    fmt
    -0.09
    Pad
    -0.08
    cljs
    -0.08
    弩
    -0.08
    åĮĸ身
    -0.08
    POSITIVE LOGITS
    .less
    0.10
    ãĥıãĥ¼ãĥī
    0.09
    溯
    0.09
    TreeWidgetItem
    0.09
    igators
    0.09
    å¨ĥ
    0.09
    TRS
    0.09
    _less
    0.09
    åĴ¬
    0.09
    _traits
    0.09
    Activations Density 0.044%

    No Known Activations