© 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. 36836
    Prev
    Next
    INDEX
    Explanations

    downtime and interruptions

    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
    漤
    -0.10
    滥
    -0.10
    æ°¡
    -0.09
     Tick
    -0.09
    è¡°
    -0.09
    缸åıį
    -0.09
     Rach
    -0.09
    追
    -0.09
    å¿ł
    -0.09
    åįij
    -0.08
    POSITIVE LOGITS
     downtime
    0.17
    åģľå·¥
    0.16
    ä¸ŃæĸŃ
    0.14
    åģľäº§
    0.14
    çĶŁäº§åĬĽ
    0.13
    çĺ«
    0.13
    æįŁå¤±
    0.12
     disruption
    0.11
    åįĬ天
    0.11
     productivity
    0.10
    Activations Density 0.129%

    No Known Activations