© 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. 27906
    Prev
    Next
    INDEX
    Explanations

    come and go

    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
     ÙĪØ®
    -0.10
    _BU
    -0.10
     Sorted
    -0.09
    æīĪ
    -0.09
    åľ¨è¿ĩ渡
    -0.09
    coming
    -0.09
    ksam
    -0.09
    ä¸Ģ声
    -0.09
     Acc
    -0.09
    ç¬ı
    -0.09
    POSITIVE LOGITS
    驶
    0.10
    è°¢
    0.10
     adjud
    0.09
    è½
    0.09
    èIJİ缩
    0.08
    ä¸Ģ座
    0.08
    å¡
    0.08
     Machines
    0.08
    æĴŃ
    0.08
    <decltype
    0.08
    Activations Density 0.028%

    No Known Activations