© 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. 10975
    Prev
    Next
    INDEX
    Explanations

    believer in

    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
    ä¸Ģ大
    -0.10
    ä¹ĭæīĢ以
    -0.09
     вне
    -0.09
     Loft
    -0.09
    æĿ¥è®²
    -0.09
    around
    -0.09
    aways
    -0.08
     ÑĦон
    -0.08
    æ½®
    -0.08
    å¾Ģå¾Ģä¼ļ
    -0.08
    POSITIVE LOGITS
    getic
    0.10
    ä»Ģä¹Īåij¢
    0.09
    usters
    0.09
    otec
    0.08
    onas
    0.08
    åĩŃçĿĢ
    0.08
    å¾Ģåīį
    0.08
    _dtype
    0.08
    顺çĿĢ
    0.08
    .step
    0.08
    Activations Density 0.011%

    No Known Activations