© 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. 30676
    Prev
    Next
    INDEX
    Explanations

    asked why

    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
     Ñģебе
    -0.09
    obb
    -0.09
    奥
    -0.08
    å¥ı
    -0.08
    举éĥ¨
    -0.08
     Ñģобой
    -0.08
    Shock
    -0.08
    çķĻç»Ļ
    -0.08
    odd
    -0.08
    digest
    -0.08
    POSITIVE LOGITS
     salute
    0.09
    ãģĦãģŁ
    0.09
    ãģĦãģªãģĦ
    0.08
     about
    0.08
     вÑģего
    0.08
    ä»Ģä¹Īæĺ¯
    0.08
    æĢĿãģ£ãģŁ
    0.08
    zier
    0.08
    åľ¨åħ¶
    0.08
     Strait
    0.08
    Activations Density 0.006%

    No Known Activations