© 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. 36307
    Prev
    Next
    INDEX
    Explanations

    reduction or "ed" verb completion

    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
    adol
    -0.10
    /from
    -0.09
    æĻļæĬ¥
    -0.09
    ãĥ¬ãĤ¹
    -0.09
    èķĬ
    -0.09
    alog
    -0.09
    åįģ足
    -0.09
    entr
    -0.09
    itation
    -0.09
    ounder
    -0.09
    POSITIVE LOGITS
     thiá»ĥu
    0.20
    åĴĮå®ĮåĸĦ
    0.12
    warf
    0.10
    çĶļèĩ³è¿ŀ
    0.10
    çĶļèĩ³
    0.10
     rÃłng
    0.10
    TexImage
    0.10
    äºĨèĩªå·±çļĦ
    0.09
    /mod
    0.09
    çĶļèĩ³è¿ĺ
    0.09
    Activations Density 0.054%

    No Known Activations