© 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. 27637
    Prev
    Next
    INDEX
    Explanations

    back

    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
    大大
    -0.10
    que
    -0.09
    dux
    -0.09
    鬻
    -0.09
    yled
    -0.09
    ç»¿åľ°
    -0.09
    çĦ¦çĤ¹
    -0.08
    å¥Ĺ
    -0.08
    çľĭåĪ°äºĨ
    -0.08
    QUE
    -0.08
    POSITIVE LOGITS
    åıįåĩ»
    0.13
    åıįæĬĹ
    0.11
    åĽŀæĬ¥
    0.10
    åĽŀé¦Ī
    0.10
    æĬµæĮ¡
    0.10
    ÙĨز
    0.09
    ress
    0.09
    /respond
    0.09
    heel
    0.09
    éĨĴ
    0.08
    Activations Density 0.058%

    No Known Activations