© 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. 5845
    Prev
    Next
    INDEX
    Explanations

    surprising enough or convenient enough

    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¸Ĭæĸ¹
    -0.10
    æ¦Ĥæĭ¬
    -0.10
    иÑĤелÑĮно
    -0.10
    è¿ĻäºĽä¸ľè¥¿
    -0.10
    éħĮ
    -0.09
    è¿Ļæł·çļĦè¯Ŀ
    -0.09
    åļ£
    -0.09
    ween
    -0.09
     authorized
    -0.09
    åĬłåĪĨ
    -0.09
    POSITIVE LOGITS
    临
    0.09
    æĵį
    0.09
    elder
    0.09
    erson
    0.09
    ãģ°ãģĭãĤĬ
    0.08
     Wiki
    0.08
    LESS
    0.08
    irm
    0.08
    åĩ¡
    0.08
    .Back
    0.08
    Activations Density 0.002%

    No Known Activations