© 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. 58561
    Prev
    Next
    INDEX
    Explanations

    opponent

    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
    綦
    -0.10
    Ùģض
    -0.10
    ä¼Ĺ
    -0.08
    ç¥ł
    -0.08
    arius
    -0.08
     Cors
    -0.08
     localization
    -0.07
    .loc
    -0.07
    åįģæľĪ
    -0.07
    еÑĢ
    -0.07
    POSITIVE LOGITS
    对æīĭ
    0.22
    对æĸ¹
    0.18
    ç«ŀäºī对æīĭ
    0.17
    ç«ŀäºī
    0.15
     opponent
    0.14
    对çŃĸ
    0.14
    对éĿ¢
    0.14
    对æĬĹ
    0.14
    æķĮ人
    0.14
    ä»ĩ
    0.14
    Activations Density 0.079%

    No Known Activations