© 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. 9517
    Prev
    Next
    INDEX
    Explanations

    just directional position

    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
    æ¯ıæ¯ı
    -0.09
    jian
    -0.09
    æĬij
    -0.09
    umd
    -0.09
     SX
    -0.09
    çħĮ
    -0.09
    IEW
    -0.09
    éĥ¤
    -0.08
     tilt
    -0.08
    éĺķ
    -0.08
    POSITIVE LOGITS
    éĿłè¿ij
    0.13
     near
    0.12
    沿
    0.10
    ä¼ļ产çĶŁ
    0.10
    éĹ¹
    0.09
    临è¿ij
    0.09
    æİ¥è¿ij
    0.09
    à¹ĥà¸ģล
    0.09
    æĸ¼
    0.09
     between
    0.09
    Activations Density 0.134%

    No Known Activations