© 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. 34018
    Prev
    Next
    INDEX
    Explanations

    welcome or welcomed word followed by nouns

    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
     Dut
    -0.10
    _normalized
    -0.08
    isphere
    -0.08
    åĨ²çªģ
    -0.08
    riding
    -0.08
    ured
    -0.08
    æ´ª
    -0.08
    驾é©Ń
    -0.08
    èªĵ
    -0.08
     GAS
    -0.08
    POSITIVE LOGITS
    欢è¿İ
    0.12
     welcome
    0.12
    ä¹ĭ举
    0.12
    欣åĸľ
    0.11
    æĮ¯å¥ĭ
    0.11
     relief
    0.10
     welcomed
    0.10
     Welcome
    0.10
    Welcome
    0.10
    é«ĺåħ´
    0.10
    Activations Density 0.025%

    No Known Activations