© 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. 56633
    Prev
    Next
    INDEX
    Explanations

    Original Message/article

    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
    æĪIJçĨŁçļĦ
    -0.09
    âĹ»
    -0.09
    æѧ
    -0.08
     Incorporated
    -0.08
    ÑģÑĤÑĢоиÑĤелÑĮ
    -0.08
    管çIJĨå·¥ä½ľ
    -0.08
    Ian
    -0.08
    åĵĻ
    -0.08
    å°Ĩèĩªå·±çļĦ
    -0.08
     Slam
    -0.08
    POSITIVE LOGITS
    IGO
    0.10
    _dispatcher
    0.09
    é¢Ħ订
    0.09
    /original
    0.09
    yg
    0.09
    Friendly
    0.09
    转载
    0.09
     Elite
    0.09
    iginal
    0.08
     оÑĢигинал
    0.08
    Activations Density 0.033%

    No Known Activations