© 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. 23641
    Prev
    Next
    INDEX
    Explanations

    larger than

    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
    lyn
    -0.10
    åķĨä¼ļ
    -0.10
    fulness
    -0.10
    çªĴ
    -0.09
    places
    -0.09
    ÙĦب
    -0.09
    croft
    -0.09
    fully
    -0.08
    èĴ»
    -0.08
     Lair
    -0.08
    POSITIVE LOGITS
    -than
    0.12
    æĽ´å¤§çļĦ
    0.12
    ï¸
    0.10
     bigger
    0.10
    itarian
    0.10
    IgnoreCase
    0.09
    æĽ´å¤§
    0.09
    mente
    0.09
    ifest
    0.08
    ieur
    0.08
    Activations Density 0.030%

    No Known Activations