© 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. 3356
    Prev
    Next
    INDEX
    Explanations

    that/who followed by verbs

    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
    æľīæīĢ
    -0.10
    缸éĢļ
    -0.09
    éĨĴ
    -0.09
    æŦ
    -0.08
    KHR
    -0.08
    pig
    -0.08
    åģļçļĦ
    -0.08
    髦
    -0.08
    _echo
    -0.08
    æĺ¯ä¸Ģ个
    -0.08
    POSITIVE LOGITS
    æİ¥å¾ħ
    0.09
    æľĢãĤĤ
    0.09
     Wag
    0.09
    åľĨå½¢
    0.09
    undy
    0.09
    iest
    0.09
     proverb
    0.09
    æľĢåħ·
    0.09
     most
    0.09
    æĬĬè¿ĻäºĽ
    0.09
    Activations Density 0.103%

    No Known Activations