© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10322
    Prev
    Next
    INDEX
    Explanations

    moving or switching from one thing to another

    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
    èŀį
    -0.10
    ä½ľçĶ¨
    -0.09
    伸åĩº
    -0.09
    edin
    -0.09
    æĭ¿åĩºæĿ¥
    -0.09
    erten
    -0.08
    交èŀį
    -0.08
    ç´§å¼ł
    -0.08
    ç»°
    -0.08
    çķ´
    -0.08
    POSITIVE LOGITS
    åĪĩæį¢
    0.16
     switching
    0.15
     switch
    0.14
    Switch
    0.14
    switch
    0.14
    å¼ĥ
    0.13
     Switch
    0.13
    è¿ģ移
    0.12
     switched
    0.12
    -switch
    0.12
    Activations Density 0.219%

    No Known Activations