© 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. 23255
    Prev
    Next
    INDEX
    Explanations

    place of safety and refuge

    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
    æĸĩèīº
    -0.09
    æŃ£éĿ¢
    -0.09
    äºļ马
    -0.09
    ÙĥاÙģ
    -0.08
    iana
    -0.08
    é²±
    -0.08
    åĽ¢éķ¿
    -0.08
    åĵĪå°Ķ
    -0.08
    æĩĤäºĭ
    -0.08
    eler
    -0.08
    POSITIVE LOGITS
     refuge
    0.19
     shelter
    0.17
    åºĩ
    0.17
    躲
    0.16
     sanctuary
    0.15
    éģ¿
    0.14
     Refuge
    0.13
     shelters
    0.13
     asylum
    0.13
     Sanctuary
    0.12
    Activations Density 0.063%

    No Known Activations