© 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. 52597
    Prev
    Next
    INDEX
    Explanations

    prison personnel and security procedures

    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
    æĭĶ
    -0.10
    å¨ĥå¨ĥ
    -0.10
    fills
    -0.09
    aber
    -0.09
    æ§İ
    -0.08
    Touch
    -0.08
    hsi
    -0.08
    æ²»æĦĪ
    -0.08
    ures
    -0.08
     sidelines
    -0.08
    POSITIVE LOGITS
    çĭ±
    0.13
    åľ¨éĩĮéĿ¢
    0.10
    åĽļ
    0.10
    å¹²èѦ
    0.09
    çĽijçĭ±
    0.09
    éªĿ
    0.09
    çįĦ
    0.09
    室
    0.09
     inmates
    0.09
    é¼IJ
    0.09
    Activations Density 0.179%

    No Known Activations