© 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. 52623
    Prev
    Next
    INDEX
    Explanations

    change the background

    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
    lez
    -0.10
    伫
    -0.08
    IRO
    -0.08
    红线
    -0.08
    ิà¸Ļ
    -0.08
    æĸ°ä¸Ģè½®
    -0.08
    竦
    -0.08
    .Modules
    -0.08
    ignment
    -0.08
     Gors
    -0.08
    POSITIVE LOGITS
    é¢ĵ
    0.10
    主æĦı
    0.10
    å·¥ä½ľä½ľé£İ
    0.09
    æı´
    0.09
    å¤īãģĪ
    0.09
    nee
    0.08
    æį¢æĪIJ
    0.08
    åīĸæŀIJ
    0.08
     fortunes
    0.08
     дело
    0.08
    Activations Density 0.098%

    No Known Activations