© 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. 27104
    Prev
    Next
    INDEX
    Explanations

    one, third, those

    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¿Ļ项
    -0.11
    该项
    -0.09
    ãģªãĤĤãģ®
    -0.09
    æѤäºĭ
    -0.09
    .cr
    -0.09
    è¿Ļ件äºĭ
    -0.09
    _hdl
    -0.09
     Schiff
    -0.09
    .CR
    -0.09
    &action
    -0.09
    POSITIVE LOGITS
    uxe
    0.08
    wish
    0.08
    é¢Ĩ
    0.08
    åīįéĿ¢
    0.08
    GIS
    0.08
    oes
    0.08
    èĤ²
    0.07
    è±ļ
    0.07
    VOKE
    0.07
     unc
    0.07
    Activations Density 0.153%

    No Known Activations