© 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. 20612
    Prev
    Next
    INDEX
    Explanations

    insist or persist

    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
    å¥ķ
    -0.09
     wen
    -0.08
    æĦ§
    -0.08
    ạn
    -0.08
    åīĥ
    -0.08
    座è°Īä¼ļä¸Ĭ
    -0.08
    æİł
    -0.07
    .detach
    -0.07
    olie
    -0.07
    Leap
    -0.07
    POSITIVE LOGITS
     insist
    0.15
    ä¸įæĩĪ
    0.13
    åĿļæĮģ
    0.13
     insistence
    0.13
    æī§æĦı
    0.13
    ä¸įèĤ¯
    0.12
     persistence
    0.12
     persisted
    0.12
     Persistence
    0.12
    ä¸Ģ缴åĪ°
    0.11
    Activations Density 0.039%

    No Known Activations