© 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. 50890
    Prev
    Next
    INDEX
    Explanations

    help me

    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
    äºijéĽĨ
    -0.09
     Ñĥдоб
    -0.09
    edList
    -0.09
     Kro
    -0.08
    .groups
    -0.08
     Karn
    -0.08
    èģĶåIJĪä¼ļ
    -0.08
    vice
    -0.08
    زÙĨ
    -0.08
    迹
    -0.08
    POSITIVE LOGITS
    æIJŀ好
    0.12
    åĬŀ好
    0.11
    åģļä¸Ģ个
    0.09
    åģļ个
    0.09
    ä¸Ģéģĵ
    0.09
    åģļ好
    0.09
    ä¸ĢåIJĮ
    0.08
    好ä¸į好
    0.08
    éĵ²
    0.08
    åĨ³ä¸į
    0.08
    Activations Density 0.025%

    No Known Activations