© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 4114
    Prev
    Next
    INDEX
    Explanations

    python scripts

    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
    è¿Ļå¼ł
    -0.09
    çı²
    -0.08
    æ´Ħ
    -0.08
    æģŃ
    -0.08
    é¹³
    -0.08
    à¸Ļาม
    -0.08
    ä¸¤å¼ł
    -0.08
    å²ļ
    -0.08
    åĶ¤
    -0.08
    éĢı
    -0.08
    POSITIVE LOGITS
    代çłģ
    0.10
     skeleton
    0.10
     code
    0.09
     skeletal
    0.09
    cents
    0.09
     код
    0.09
    é¢Ĩåıĸ
    0.09
    使æĪij
    0.08
     gén
    0.08
    示æĦı
    0.08
    Activations Density 1.406%

    No Known Activations