© 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. 21272
    Prev
    Next
    INDEX
    Explanations

    retinal occlusion

    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
    åľĪ
    -0.10
     ìłijê·¼
    -0.10
    Past
    -0.09
    京津
    -0.09
    .Touch
    -0.09
    çįŃ
    -0.09
    ä¸įå¼Ģ
    -0.09
    ocl
    -0.08
    两åįĥ
    -0.08
    kir
    -0.08
    POSITIVE LOGITS
    èµĭ
    0.09
    ä»İä¸ļ
    0.09
    缺
    0.09
    ulis
    0.08
     Standing
    0.08
    ping
    0.08
    ä¸Ńæĸ°
    0.08
    积水
    0.08
    (Duration
    0.08
    æ³Ľ
    0.08
    Activations Density 0.255%

    No Known Activations