© 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. 1299
    Prev
    Next
    INDEX
    Explanations

    bridge the gap

    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
    æľ¬ç½ij
    -0.10
    udas
    -0.09
    ä½İä»·
    -0.09
     UNIQUE
    -0.08
    undy
    -0.08
    æĿĤ
    -0.08
    asts
    -0.08
    âĹĮ
    -0.08
    hands
    -0.08
    untas
    -0.08
    POSITIVE LOGITS
     gap
    0.25
    gap
    0.23
    å·®è·Ŀ
    0.23
    Gap
    0.23
    -gap
    0.21
     Gap
    0.20
     gaps
    0.19
    缺åı£
    0.18
    _gap
    0.16
    éĹ´éļĻ
    0.16
    Activations Density 0.050%

    No Known Activations