© 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. 23165
    Prev
    Next
    INDEX
    Explanations

    crime and murder cases

    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
    觥
    -0.11
    黾
    -0.10
    å¼ģ
    -0.10
    ulton
    -0.10
     Clem
    -0.10
    ä¹¾åĿ¤
    -0.10
    åĿŀ
    -0.10
    庸
    -0.09
    ä½ľèĢħæľ¬äºº
    -0.09
    ä¸Ńåįİ
    -0.09
    POSITIVE LOGITS
    æ¡Ī件
    0.14
     cases
    0.12
    侦
    0.11
    æѤæ¡Ī
    0.11
    èĬ±çº¹
    0.10
    å©ļåIJİ
    0.10
     crime
    0.10
     murders
    0.09
    AME
    0.09
    chet
    0.09
    Activations Density 0.120%

    No Known Activations