© 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. 58831
    Prev
    Next
    INDEX
    Explanations

    sand and sandbox

    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
    ä¹Ŀå·ŀ
    -0.09
    Congress
    -0.09
    flip
    -0.09
    æijĦ
    -0.08
    itto
    -0.08
     Mix
    -0.08
    pler
    -0.08
     CFR
    -0.08
     coincidence
    -0.08
    coin
    -0.08
    POSITIVE LOGITS
    æ²Ļ
    0.17
     Sand
    0.13
     sand
    0.13
     sandbox
    0.12
    sandbox
    0.12
     пеÑģ
    0.11
     Sandbox
    0.11
    Sand
    0.11
    çłĤ
    0.11
    æ²Ļ滩
    0.10
    Activations Density 0.056%

    No Known Activations