© 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. 8656
    Prev
    Next
    INDEX
    Explanations

    did

    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°±ä¼ļ
    -0.14
    便ä¼ļ
    -0.13
    ä¹Łä¼ļ
    -0.13
    å·²äºİ
    -0.12
    æľĥ
    -0.12
     would
    -0.12
    åı¯èĥ½ä¼ļ
    -0.12
    å°Ĩä¼ļ
    -0.12
    ä¼ļå¼ķèµ·
    -0.12
    would
    -0.12
    POSITIVE LOGITS
    顺åĪ©
    0.10
    æĪIJ为ä¸Ģ个
    0.09
    红åĪ©
    0.09
    æĪIJè´¥
    0.09
    ä¸įäºĨ
    0.09
    æŁIJ个
    0.09
    inus
    0.09
    åıĺæĪIJ
    0.09
    羡
    0.09
    驾
    0.09
    Activations Density 0.125%

    No Known Activations