© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51468
    Prev
    Next
    INDEX
    Explanations

    art form, mastered, grasp

    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
    ular
    -0.09
    ÃŃv
    -0.09
    溱
    -0.09
    çŀİ
    -0.08
     cum
    -0.08
     Mos
    -0.08
    _kb
    -0.08
    buch
    -0.08
    ä¸Ģåħ±
    -0.08
    ger
    -0.08
    POSITIVE LOGITS
    æİĮæı¡
    0.13
    æİĮæı¡äºĨ
    0.12
    æŀ¢
    0.10
     mastered
    0.10
    æ³¼
    0.09
    åģļåĪ°
    0.09
    å°ıäºĭ
    0.09
    åѦä¼ļäºĨ
    0.09
    æ½ĩ
    0.08
    åĮºåĨħ
    0.08
    Activations Density 0.028%

    No Known Activations