© 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. 31612
    Prev
    Next
    INDEX
    Explanations

    within text boundaries

    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
    edes
    -0.10
    妫
    -0.10
    oth
    -0.09
    iem
    -0.09
    aat
    -0.09
    åĪ«çļĦ
    -0.09
    onian
    -0.09
     ведÑĮ
    -0.09
    arb
    -0.08
     когда
    -0.08
    POSITIVE LOGITS
    æ·±å¤Ħ
    0.12
    /out
    0.12
    çĶŁäº§æĢ»åĢ¼
    0.10
    /Internal
    0.09
    ernet
    0.09
    ä¸į说
    0.09
    æ½®æµģ
    0.09
    çĸļ
    0.09
    ated
    0.09
    çļĦä¸ĢåĪĩ
    0.09
    Activations Density 0.055%

    No Known Activations