© 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. 39159
    Prev
    Next
    INDEX
    Explanations

    ArcGIS

    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
    稿
    -0.09
    ãģĹãģ¾ãģĨ
    -0.08
    rat
    -0.08
    Ŀå§ĭ
    -0.08
    åľ©
    -0.08
    oral
    -0.08
    apis
    -0.08
    عÙĤ
    -0.08
    ApiClient
    -0.07
    ateral
    -0.07
    POSITIVE LOGITS
    弧
    0.13
    uate
    0.12
    aic
    0.12
    (arc
    0.11
    itecture
    0.10
    arine
    0.10
     arc
    0.10
     Arc
    0.09
    gis
    0.09
    ulary
    0.09
    Activations Density 0.019%

    No Known Activations