© 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. 30210
    Prev
    Next
    INDEX
    Explanations

    water and related contexts

    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
    åħ¶æīĢ
    -0.10
    à¸Ńาหาร
    -0.10
    antu
    -0.09
    æīĭæ³ķ
    -0.09
     juices
    -0.09
    们
    -0.09
     dame
    -0.09
    aml
    -0.09
    çıŃ
    -0.08
    alin
    -0.08
    POSITIVE LOGITS
    èݲ
    0.11
    çıł
    0.10
    饺
    0.10
    melon
    0.10
    è¿Ľåĩº
    0.09
    樽
    0.09
    iele
    0.09
    é¾Ļ头
    0.09
    èį·èĬ±
    0.09
    èݲèĬ±
    0.09
    Activations Density 0.041%

    No Known Activations