© 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. 61500
    Prev
    Next
    INDEX
    Explanations

    gentile / gent, gently / m, , gentiles / , gentlemen

    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
    æŶåħī
    -0.10
    éĵĽ
    -0.09
    æĬķ注
    -0.09
    æŬ
    -0.09
    èĿĮ
    -0.09
    çĻ¾åĪĨçĻ¾
    -0.09
    éĶµ
    -0.08
    æĹłä¸į
    -0.08
    æĺ¥å¤ı
    -0.08
    izu
    -0.08
    POSITIVE LOGITS
    æIJŃ
    0.10
    èĤļ
    0.10
     Colleg
    0.09
    rete
    0.08
    asca
    0.08
    abox
    0.08
    utto
    0.08
    uce
    0.08
    æķ´ä¸ªäºº
    0.08
    аÑĢ
    0.08
    Activations Density 0.005%

    No Known Activations