© 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. 6373
    Prev
    Next
    INDEX
    Explanations

    piece of

    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
    IMATE
    -0.09
    è¿Ļä¸ĢåĪĩ
    -0.09
    ä¸ĢåĪĩéĥ½
    -0.09
    è¡·å¿ĥ
    -0.09
     полÑĥÑĩа
    -0.08
    ä¸Ģ件äºĭ
    -0.08
    ãģ¤ãģĦ
    -0.08
    ghan
    -0.08
    éĴ³
    -0.08
    -reply
    -0.08
    POSITIVE LOGITS
     software
    0.15
    软件
    0.15
     equipment
    0.13
    software
    0.11
    设å¤ĩ
    0.10
     обоÑĢÑĥдование
    0.10
    ç²īç¢İ
    0.10
    equipment
    0.10
    代çłģ
    0.09
    è»Łé«Ķ
    0.09
    Activations Density 0.351%

    No Known Activations