© 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. 48664
    Prev
    Next
    INDEX
    Explanations

    illegal possession

    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
     willingness
    -0.10
    é¢Ħæ¡Ī
    -0.09
    è¡°
    -0.09
     dying
    -0.09
    ä¸įåIJĥ
    -0.09
    ä¸įåıĬ
    -0.09
    æ³µ
    -0.09
    å¾ĩ
    -0.08
     assassination
    -0.08
    æĬ½æŁ¥
    -0.08
    POSITIVE LOGITS
     possession
    0.20
    åįłæľī
    0.18
    æĭ¥æľī
    0.15
     possess
    0.14
    æĭ¥
    0.13
     custody
    0.13
    æĮģæľī
    0.13
     sợ
    0.12
    æīĭä¸Ń
    0.12
    æĵģæľī
    0.12
    Activations Density 0.068%

    No Known Activations