© 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. 27930
    Prev
    Next
    INDEX
    Explanations

    contradictory or different stances

    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
    æ³Ľ
    -0.11
    è´¨
    -0.10
     Ùħد
    -0.09
    質
    -0.09
    ä¸įåħ¬å¹³
    -0.09
    æĪIJåĬŁ
    -0.09
     rightful
    -0.08
    eni
    -0.08
    çļĦæĥħ
    -0.08
    Ùħد
    -0.08
    POSITIVE LOGITS
    ç«ĭåľº
    0.14
    åı£å¾Ħ
    0.11
    æĬķ票
    0.10
    æĹĹå¸ľ
    0.09
    代è¨Ģ
    0.09
    èĪµ
    0.09
     alignment
    0.09
    ç«ŀéĢī
    0.09
    表达
    0.09
    éī´åĪ«
    0.09
    Activations Density 0.092%

    No Known Activations