© 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. 1851
    Prev
    Next
    INDEX
    Explanations

    after "the" capitalized

    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
    åĪĨå±Ģ
    -0.08
     mote
    -0.08
    满äºĨ
    -0.08
    ãģ¨ãģĦãģ£ãģ¦
    -0.08
    æľīäºĨ
    -0.08
    鸱
    -0.08
    million
    -0.08
     stead
    -0.08
    å°ij许
    -0.08
    ä»İä¸Ń
    -0.08
    POSITIVE LOGITS
    é¢ĨåĨĽ
    0.10
    é¢Ħè§ģ
    0.09
    SSIP
    0.09
    æģ¤
    0.09
     ÐijÑĥд
    0.09
     اÙĦبÙĬاÙĨات
    0.09
    alion
    0.09
    åıĹåĪ°äºĨ
    0.09
     Meaning
    0.09
    ephy
    0.08
    Activations Density 0.131%

    No Known Activations