© 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. 63494
    Prev
    Next
    INDEX
    Explanations

    guardian, keeper, custodian

    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
    éĢı
    -0.09
    од
    -0.09
    idente
    -0.09
    è¿¥
    -0.09
    IFF
    -0.08
    æĭĽåij¼
    -0.08
    ific
    -0.08
     Thr
    -0.08
    æĻ°
    -0.08
    l
    -0.08
    POSITIVE LOGITS
    è¯ł
    0.11
    ä¸ĩ家
    0.11
    æľŁåĨħ
    0.10
    keeping
    0.10
    keepers
    0.10
     cust
    0.09
    ITOR
    0.09
     custody
    0.09
    ä¸įä½ı
    0.09
    绿水
    0.09
    Activations Density 0.033%

    No Known Activations