© 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. 8012
    Prev
    Next
    INDEX
    Explanations

    agreement and disagreement

    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
    ieder
    -0.10
    æīĺ管
    -0.10
    ROL
    -0.10
     Guar
    -0.09
    ç¯ĩ
    -0.09
    æĭIJ
    -0.09
    DMI
    -0.09
    éĢłè¡Ģ
    -0.09
     Trail
    -0.08
     PTR
    -0.08
    POSITIVE LOGITS
    èµŀåIJĮ
    0.19
    ä¸įåIJĮæĦı
    0.16
    åIJĮæĦı
    0.16
     agree
    0.15
     agrees
    0.14
    agree
    0.13
     Agree
    0.13
     agreement
    0.13
     agreeing
    0.13
    认åIJĮ
    0.12
    Activations Density 0.043%

    No Known Activations