© 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. 17856
    Prev
    Next
    INDEX
    Explanations

    types of activities

    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
    åį¡éĢļ
    -0.11
    æĪij们认为
    -0.09
    ãģŀ
    -0.09
    æĭ¿èµ·
    -0.08
    æģIJæħĮ
    -0.08
    âĿIJ
    -0.08
    æ¢Ń
    -0.08
    awah
    -0.08
    hots
    -0.08
    hir
    -0.08
    POSITIVE LOGITS
     wash
    0.09
    å®ŀéĻħè¡ĮåĬ¨
    0.09
    pret
    0.09
    åħ¬ä¼Ĺ
    0.09
    çİ°åľº
    0.09
    aging
    0.08
    éĤĢ请
    0.08
    ML
    0.08
     pier
    0.08
     decay
    0.08
    Activations Density 0.052%

    No Known Activations