© 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. 56707
    Prev
    Next
    INDEX
    Explanations

    names ending in 'er' or 'ing'

    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
    ä½Ĩ对äºİ
    -0.09
    stre
    -0.09
    ä½Ĩå¦Ĥæŀľ
    -0.09
    ä½Ĩå®ĥ
    -0.09
    äºĨä¸Ģåı¥
    -0.09
    aginator
    -0.09
    ä¼Ļ
    -0.09
    té
    -0.09
    社ä¼ļåIJĦçķĮ
    -0.08
    æĥķ
    -0.08
    POSITIVE LOGITS
     dozen
    0.10
    cape
    0.08
    ç»
    0.08
    arios
    0.08
    epad
    0.08
    æĭIJ
    0.08
    UCH
    0.08
    SED
    0.08
    imps
    0.08
     crest
    0.07
    Activations Density 1.786%

    No Known Activations