© 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. 34981
    Prev
    Next
    INDEX
    Explanations

    batter, patter

    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
    ../../../
    -0.15
    ../../
    -0.11
    inke
    -0.09
    usal
    -0.09
    orris
    -0.09
    ce
    -0.09
    ../../../../
    -0.08
    elts
    -0.08
    atum
    -0.08
    sm
    -0.08
    POSITIVE LOGITS
    oons
    0.10
    lasses
    0.10
    ÑĢон
    0.10
    æĪIJåŀĭ
    0.10
    çĦ¶æĺ¯
    0.09
    ments
    0.09
    ment
    0.09
    ween
    0.09
    kö
    0.09
    ÙģÙĤ
    0.09
    Activations Density 0.003%

    No Known Activations