© 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. 31752
    Prev
    Next
    INDEX
    Explanations

    leads to death or damage

    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
    èĪ¶
    -0.10
    é¢ĩæľī
    -0.09
    æľīæĦı
    -0.09
    墨
    -0.09
    åıĹ欢è¿İ
    -0.08
    èł¢
    -0.08
    çķ¥æľī
    -0.08
    空åīį
    -0.08
    代çIJĨ人
    -0.08
    .wp
    -0.08
    POSITIVE LOGITS
    æľĢç»Ī
    0.13
    æľĢçµĤ
    0.12
     eventually
    0.11
     Eventually
    0.11
    Eventually
    0.11
     eventual
    0.10
    æĢ§åij½
    0.10
    çĶļèĩ³
    0.10
    åį±
    0.10
    è¿ĽèĢĮ
    0.10
    Activations Density 0.338%

    No Known Activations