© 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. 6595
    Prev
    Next
    INDEX
    Explanations

    confectionery concoction conglomerate decoction conundrum

    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
    fully
    -0.09
    ussy
    -0.09
    åĹŁ
    -0.09
    dden
    -0.09
    elter
    -0.09
    pecia
    -0.08
    dif
    -0.08
    åĮ¡
    -0.08
    prung
    -0.08
    ache
    -0.08
    POSITIVE LOGITS
    สรร
    0.10
    ivity
    0.10
    openhagen
    0.09
    sm
    0.09
    ery
    0.08
    try
    0.08
    ãĥĵãĤ¸
    0.08
    Priv
    0.08
    ment
    0.08
    urable
    0.08
    Activations Density 0.006%

    No Known Activations