© 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. 31838
    Prev
    Next
    INDEX
    Explanations

    guilty of

    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
    åı£ä¸Ń
    -0.09
     ÑģÑĤоим
    -0.09
    å¼ģ
    -0.09
    åıĭåĸĦ
    -0.09
     мÑıгк
    -0.08
    çĶŁçĹħ
    -0.08
    èĥ½åĬĽå¼º
    -0.08
    íĮ©
    -0.08
    åĵģæł¼
    -0.08
     ÑĢабоÑĤник
    -0.08
    POSITIVE LOGITS
     guilty
    0.10
    asso
    0.09
    /inc
    0.09
    ãģ¡ãģ¯
    0.09
    /lgpl
    0.08
    交
    0.08
    累累
    0.08
    ona
    0.08
    adera
    0.08
    ties
    0.08
    Activations Density 0.015%

    No Known Activations