© 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. 33130
    Prev
    Next
    INDEX
    Explanations

    best known

    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
    ä¸įå°ı
    -0.09
    念头
    -0.09
    æĻ¯å¾·
    -0.09
    ØŃÙĥ
    -0.09
    å¿ĥåºķ
    -0.08
    ä»Ģä¹Īæł·
    -0.08
    æĢĢæĬ±
    -0.08
    ç²
    -0.08
    lds
    -0.08
    ographers
    -0.08
    POSITIVE LOGITS
    为广大
    0.10
    -known
    0.10
    对åѦçĶŁ
    0.09
    -quality
    0.09
    aland
    0.09
     repl
    0.09
    åı°é£İ
    0.09
    à¸Ľà¸ı
    0.09
    è¡Ķæİ¥
    0.08
    /fast
    0.08
    Activations Density 0.046%

    No Known Activations