© 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. 4235
    Prev
    Next
    INDEX
    Explanations

    expressing opinions or ideas

    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
     пÑĢедÑĥÑģмоÑĤÑĢ
    -0.09
    _portfolio
    -0.09
    ä¼ı
    -0.09
    гÑĥ
    -0.08
    çĹħ
    -0.08
    å±ij
    -0.08
    çĹĤ
    -0.08
    åħ¸åŀĭçļĦ
    -0.08
     fearing
    -0.08
    _detach
    -0.08
    POSITIVE LOGITS
    表达
    0.18
    表达äºĨ
    0.16
    äºĨèĩªå·±çļĦ
    0.16
    æĬĬèĩªå·±çļĦ
    0.15
    èĩªå·±çļĦ
    0.15
    对èĩªå·±çļĦ
    0.14
    Express
    0.13
    express
    0.13
     expressing
    0.13
    ä¼łè¾¾
    0.13
    Activations Density 0.071%

    No Known Activations