© 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. 34215
    Prev
    Next
    INDEX
    Explanations

    reporting verbs followed by clauses

    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
    è¿Ľåİ»
    -0.10
    好åIJĹ
    -0.10
    ä½łèĩªå·±
    -0.10
    为ä¹ĭ
    -0.09
    ä»ĸèĩªå·±
    -0.09
    ãģ¨ãģ¯ãģĦãģĪ
    -0.09
    æīĢ说çļĦ
    -0.09
    羣çļĦ
    -0.09
    èĢ¨
    -0.09
    æľķ
    -0.09
    POSITIVE LOGITS
    主è¦ģæľī
    0.09
    AccessType
    0.09
    forcer
    0.08
    å°¤åħ¶
    0.08
    Çİ
    0.08
    .mc
    0.08
    /general
    0.08
    ismic
    0.08
     decisive
    0.07
     Yue
    0.07
    Activations Density 0.083%

    No Known Activations