© 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. 5510
    Prev
    Next
    INDEX
    Explanations

    connect four

    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
    usk
    -0.10
    åĪ·
    -0.10
    pdev
    -0.09
    é¢Īéĥ¨
    -0.09
    èĩªè´Ł
    -0.09
    åĴĮçĿ¦
    -0.09
    ãĥªãĥ¼
    -0.08
    çIJĨäºĭä¼ļ
    -0.08
    usal
    -0.08
    åĬ²
    -0.08
    POSITIVE LOGITS
    çŃ¹çłģ
    0.15
    骰
    0.12
    èĬ±æľµ
    0.10
    æ£ĭ
    0.09
    çļĦé¢ľèī²
    0.09
    é»ijçĻ½
    0.09
    çīĮ
    0.09
    æĪı
    0.09
    ä¸Ģæŀļ
    0.09
    çģµéŃĤ
    0.09
    Activations Density 0.075%

    No Known Activations