© 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. 39764
    Prev
    Next
    INDEX
    Explanations

    reject, refuse, deny

    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
    æĤ¼
    -0.10
    ä¸İä¼Ĺ
    -0.10
     ÑħоÑĤÑĮ
    -0.09
    VERS
    -0.08
    .CopyTo
    -0.08
    aptor
    -0.08
     yên
    -0.08
    æī¬
    -0.08
    IRD
    -0.08
    å°Ĭ
    -0.08
    POSITIVE LOGITS
    æĭĴç»Ŀ
    0.28
    æĭĴ
    0.26
     reject
    0.22
    æĭĴçµķ
    0.21
     Reject
    0.21
     rejection
    0.20
     rejected
    0.20
    Reject
    0.20
     rejects
    0.20
    reject
    0.20
    Activations Density 0.091%

    No Known Activations