© 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. 27232
    Prev
    Next
    INDEX
    Explanations

    actioning words

    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
    è¿Ļä¹Łæĺ¯
    -0.10
    mant
    -0.09
    ops
    -0.09
    ÑĢеÑģ
    -0.09
    æģ¯
    -0.08
    åģ¥åº·çļĦ
    -0.08
    导读
    -0.08
    è¿Ļæĺ¯æĪij们
    -0.08
    æijĨ
    -0.08
     hal
    -0.08
    POSITIVE LOGITS
    磺
    0.09
    üz
    0.09
    hattan
    0.09
    ilers
    0.09
    .unregister
    0.09
    dez
    0.09
    nex
    0.08
    é¢ĺ
    0.08
    alla
    0.08
    indo
    0.08
    Activations Density 0.051%

    No Known Activations