© 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. 55969
    Prev
    Next
    INDEX
    Explanations

    congressman or representative names and districts

    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
    opsis
    -0.10
    èĵ¬
    -0.10
    UES
    -0.10
    ëĭĿ
    -0.09
    ILLED
    -0.09
     Buccane
    -0.09
    osas
    -0.09
    sonian
    -0.09
    èIJ½ä¸ĭ
    -0.09
    宽æĿ¾
    -0.09
    POSITIVE LOGITS
    ally
    0.10
    aint
    0.09
    æľ¬ç«Ļ
    0.08
    ively
    0.08
     unheard
    0.08
    jin
    0.08
     Brill
    0.08
    ä»¥äººä¸ºæľ¬
    0.08
    ighth
    0.08
    eur
    0.08
    Activations Density 0.010%

    No Known Activations