© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50545
    Prev
    Next
    INDEX
    Explanations

    joining 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
    é«
    -0.08
    лей
    -0.08
    让人们
    -0.08
    ifi
    -0.08
    æī«é»ij
    -0.08
    take
    -0.08
    ãģ¨ãģĦãģ£ãģŁ
    -0.08
    触åıĬ
    -0.08
    鸥
    -0.08
    认æ¸ħ
    -0.08
    POSITIVE LOGITS
    ervices
    0.09
    erg
    0.09
    åħ¼
    0.09
    éĺ»
    0.08
    æļ¨
    0.08
    ового
    0.08
    its
    0.08
    oret
    0.08
    æĪĸèĢħ说
    0.08
    lag
    0.08
    Activations Density 0.114%

    No Known Activations