© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 59372
    Prev
    Next
    INDEX
    Explanations

    Korean and South Korean contexts

    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
    Negative Logits
    enga
    -0.09
    homes
    -0.09
     Leap
    -0.08
    ErrorHandler
    -0.08
    å®ģåİ¿
    -0.08
     Clause
    -0.08
    éĿ¡
    -0.08
    åıĹå½±åĵį
    -0.08
    íľ´
    -0.08
    istic
    -0.08
    POSITIVE LOGITS
    æĺ¯éŁ©åĽ½
    0.17
     Seoul
    0.13
     Korean
    0.13
     Koreans
    0.12
    éŁ©åĽ½
    0.12
     Korea
    0.11
    æľĿé²ľ
    0.11
     Pyongyang
    0.10
    æľ´
    0.10
     HÃłn
    0.10
    Activations Density 0.028%

    No Known Activations