© 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. 19309
    Prev
    Next
    INDEX
    Explanations

    believe, trust, own

    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
    ç»Ĭ
    -0.09
     rejection
    -0.09
    汤
    -0.09
     wed
    -0.08
    è¸ı
    -0.08
    çµIJ
    -0.08
    ä¸įå¼Ģ
    -0.08
    udas
    -0.08
    egin
    -0.08
    _shadow
    -0.08
    POSITIVE LOGITS
    宣称
    0.12
    声称
    0.11
     ÙĤÙĪÙĦÙĩ
    0.10
    äºĽä»Ģä¹Ī
    0.09
    说çļĦè¯Ŀ
    0.09
    å®ĺ宣
    0.09
    ç¾İåĮĸ
    0.09
    èĩªç§°
    0.09
    åĩłåı¥
    0.09
    éĤ£äºĽ
    0.08
    Activations Density 0.044%

    No Known Activations