© 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. 50168
    Prev
    Next
    INDEX
    Explanations

    false or misleading statements, serious injury, torture

    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
    ç®IJ
    -0.09
    ====
    -0.08
    çĻŀ
    -0.08
    é«ĺè¾¾
    -0.07
    æ³IJ
    -0.07
    纺
    -0.07
    æĭīåĬ¨
    -0.07
     chá»ĵng
    -0.07
    åĽĽå¤Ħ
    -0.07
    lene
    -0.07
    POSITIVE LOGITS
    ibri
    0.10
     Hữu
    0.09
    ÑĭÑĤа
    0.09
    æĪĸèĢħåħ¶ä»ĸ
    0.09
    orts
    0.09
    enity
    0.09
    åŃª
    0.09
    è¿ijæľŁ
    0.08
    åĺĽ
    0.08
    ospace
    0.08
    Activations Density 0.108%

    No Known Activations