© 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. 39758
    Prev
    Next
    INDEX
    Explanations

    unnecessary negative consequences

    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
    ä¸įåIJĪæł¼
    -0.09
    çªģåĩ»
    -0.09
    ATIC
    -0.09
    è¦ģéĹ»
    -0.09
    åıªä¸įè¿ĩ
    -0.09
    å¿ĺè®°äºĨ
    -0.08
    åħ³äºİæĪij们
    -0.08
    ä¸įåĩĨ
    -0.08
    å¿ĺè®°
    -0.08
    .Man
    -0.08
    POSITIVE LOGITS
    ä¸įå¿ħè¦ģçļĦ
    0.12
    麻çĥ¦
    0.09
    çĸ¯
    0.09
    ç¼ķ
    0.09
    æĿŀ
    0.09
    ément
    0.09
     fr
    0.09
     mÃŃ
    0.09
    ENO
    0.08
     bloody
    0.08
    Activations Density 0.018%

    No Known Activations