© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7955
    Prev
    Next
    INDEX
    Explanations

    watch or see

    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
    ä¿Ŀå®Ī
    -0.09
    åĭĻ
    -0.09
     cud
    -0.09
    缶
    -0.09
    å²±
    -0.09
     Ded
    -0.09
    åĬ¡
    -0.09
    åĢ¾
    -0.09
    ä¸Ģå¥Ĺ
    -0.09
    older
    -0.08
    POSITIVE LOGITS
    зн
    0.09
     ТÑĥÑĢ
    0.09
    æı´
    0.08
    è£Ļ
    0.08
    ees
    0.08
    社交
    0.08
    MBER
    0.08
    社
    0.08
    书æĪ¿
    0.08
    mil
    0.08
    Activations Density 0.009%

    No Known Activations