© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52677
    Prev
    Next
    INDEX
    Explanations

    inflammation, irritation, disruption

    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
    æķĪæŀľ
    -0.09
    件
    -0.09
    æīij
    -0.09
    èħ®
    -0.08
    eph
    -0.08
    夯å®ŀ
    -0.08
    åıĤä¸İèĢħ
    -0.08
    oland
    -0.08
    建档
    -0.08
    æħİéĩį
    -0.08
    POSITIVE LOGITS
    åºĵéĩĮ
    0.10
    主è¦ģåİŁåĽł
    0.09
    trs
    0.09
    кли
    0.09
     either
    0.09
    太è¿ĩ
    0.08
    æģ£
    0.08
    سب
    0.08
     Either
    0.08
    绶
    0.08
    Activations Density 0.465%

    No Known Activations