© 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. 53113
    Prev
    Next
    INDEX
    Explanations

    o following or preceding other characters

    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
    ault
    -0.10
    l
    -0.10
    ous
    -0.09
    家ä½ı
    -0.09
    s
    -0.09
    ç»Ŀ大å¤ļæķ°
    -0.09
    ãģıãĤĭ
    -0.08
    k
    -0.08
    n
    -0.08
    æļĤåģľ
    -0.08
    POSITIVE LOGITS
    xygen
    0.12
    yssey
    0.11
    scar
    0.11
    portunity
    0.10
    IGINAL
    0.10
    'clock
    0.10
    cean
    0.10
    essional
    0.10
    portun
    0.10
    onas
    0.10
    Activations Density 0.061%

    No Known Activations