© 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. 34027
    Prev
    Next
    INDEX
    Explanations

    hard to or hard for

    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
    adio
    -0.09
     Valent
    -0.08
    èĩĥ
    -0.08
     Commons
    -0.08
    è¿ĻäºĽéĹ®é¢ĺ
    -0.08
    resolution
    -0.08
    sembl
    -0.08
    *)&
    -0.08
    ä¸ĩä¸Ģ
    -0.08
    åĪĨ辨çİĩ
    -0.07
    POSITIVE LOGITS
    æ¯ı个人çļĦ
    0.10
    æľī人说
    0.10
    说æĺ¯
    0.10
    让人
    0.09
    enaire
    0.09
    éĶ»çĤ¼
    0.09
    人çļĦ
    0.09
    è¿Ļæł·è¯´
    0.09
    èĩ
    0.09
    _soft
    0.09
    Activations Density 0.017%

    No Known Activations