© 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. 48002
    Prev
    Next
    INDEX
    Explanations

    playing or acting

    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
    alias
    -0.09
    .bulk
    -0.09
    áºŃp
    -0.09
    à¸Ĺราà¸ļ
    -0.09
     UPC
    -0.09
    .alias
    -0.09
    天çĦ¶
    -0.09
    åħĪ天
    -0.09
    家ä¼Ļ
    -0.08
    rello
    -0.08
    POSITIVE LOGITS
    æĻºåĬĽ
    0.09
    åł¡
    0.09
     muster
    0.09
    สà¸Ļ
    0.08
     Barb
    0.08
    ä¸į对
    0.08
    æī¿æĭħ
    0.08
    ä¸ĺ
    0.08
     fend
    0.08
    yu
    0.08
    Activations Density 0.028%

    No Known Activations