© 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. 9774
    Prev
    Next
    INDEX
    Explanations

    marker studios, pens, images

    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
    åħ¨èµĦ
    -0.10
    ning
    -0.09
    mask
    -0.09
     одно
    -0.09
    devil
    -0.08
    esty
    -0.08
    icut
    -0.08
    andy
    -0.08
    æ´®
    -0.08
    两æīĭ
    -0.08
    POSITIVE LOGITS
    eros
    0.10
    å¿Ŀ
    0.09
    eras
    0.09
    ç¾Ķ
    0.09
    opot
    0.09
    çŃı
    0.09
    otle
    0.08
    upo
    0.08
    æľīä¸Ģ次
    0.08
     sore
    0.08
    Activations Density 0.013%

    No Known Activations