© 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. 32651
    Prev
    Next
    INDEX
    Explanations

    lucky winner, break, old, enough

    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
    ãģ£ãģįãĤĬ
    -0.12
    ives
    -0.10
    r
    -0.09
    icular
    -0.09
    ding
    -0.09
    idable
    -0.09
    cpy
    -0.09
    大å±Ģ
    -0.09
    ulate
    -0.08
    åı¯èĥ½æĢ§
    -0.08
    POSITIVE LOGITS
    æĿ¼
    0.10
    AMA
    0.09
    ousy
    0.09
    اÙħÙĦ
    0.09
    hea
    0.09
    æĥķ
    0.09
    åĿ¤
    0.09
     ÃŃch
    0.09
    PIO
    0.09
    NESS
    0.09
    Activations Density 0.008%

    No Known Activations