© 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. 34691
    Prev
    Next
    INDEX
    Explanations

    extracting text

    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
    稳妥
    -0.09
    qli
    -0.08
    otti
    -0.08
    -fw
    -0.08
    æªĢ
    -0.08
    ä¹ĭä½ľ
    -0.08
    IRT
    -0.08
     Toll
    -0.08
    æ¯ĶåĪ©
    -0.08
    Nib
    -0.07
    POSITIVE LOGITS
    .TXT
    0.12
    é¢ľ
    0.10
    /text
    0.10
    æĸĩæľ¬
    0.10
    /plain
    0.10
    =text
    0.10
    éĩijåŃĹå¡Ķ
    0.09
    é¡ı
    0.09
    itarian
    0.09
    uality
    0.09
    Activations Density 0.159%

    No Known Activations