© 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. 16680
    Prev
    Next
    INDEX
    Explanations

    steps, stairs, or staircases

    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
    ttp
    -0.11
    çļĦå®ŀåĬĽ
    -0.10
    ׾×Ļ×Ŀ
    -0.09
    è¿ģ
    -0.09
    éļĻ
    -0.09
    端æŃ£
    -0.09
    é¤į
    -0.09
    elsing
    -0.09
    å®ŀåĬĽ
    -0.09
    ìķ¡
    -0.09
    POSITIVE LOGITS
    åı°éĺ¶
    0.17
     stairs
    0.12
    steps
    0.11
    楼梯
    0.10
    éĺ¶
    0.10
     steps
    0.10
    ord
    0.10
    éĺ¶æ¢¯
    0.09
     Tender
    0.09
    è¶ĭ
    0.08
    Activations Density 0.013%

    No Known Activations