© 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. 2683
    Prev
    Next
    INDEX
    Explanations

    fairly + adjective

    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
    oretical
    -0.10
    vant
    -0.09
    ا
    -0.09
    ption
    -0.09
    oub
    -0.09
    ive
    -0.08
    éĶĻäºĨ
    -0.08
     thÆ°á»Ľc
    -0.08
    ore
    -0.08
    大楼
    -0.08
    POSITIVE LOGITS
    举åĬ¨
    0.10
    istically
    0.10
     обÑĢазом
    0.10
    ä¸İæѤ
    0.09
    ách
    0.09
    ities
    0.09
    .complete
    0.09
    èħ¼
    0.09
     dozen
    0.09
    ucchini
    0.09
    Activations Density 0.005%

    No Known Activations