© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51426
    Prev
    Next
    INDEX
    Explanations

    political parties

    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
    æIJª
    -0.10
    venge
    -0.09
    ä¼łéĶĢ
    -0.09
    quip
    -0.09
    ä¸įç»ıæĦı
    -0.09
    ä¸įåħ¬å¹³
    -0.08
     dro
    -0.08
    ä¸ŃåĽ½ä¼ģä¸ļ
    -0.08
    åıĮèĦļ
    -0.08
    زار
    -0.08
    POSITIVE LOGITS
    æī§æĶ¿
    0.10
     cánh
    0.10
    _party
    0.10
     Parties
    0.10
    æ©Ļ
    0.09
    AINS
    0.09
    åħļåĴĮ
    0.09
    åħļ
    0.09
     party
    0.09
    èģĶ缣
    0.09
    Activations Density 0.143%

    No Known Activations