© 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. 17170
    Prev
    Next
    INDEX
    Explanations

    Gill, Peyton, or names ending in Gill

    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
    å®ļäºĨ
    -0.13
    ess
    -0.11
    æĹ¶ç©º
    -0.11
    haft
    -0.10
     возд
    -0.09
    累累
    -0.09
    ikon
    -0.09
    ives
    -0.09
    alom
    -0.09
     Jos
    -0.08
    POSITIVE LOGITS
    ery
    0.10
    room
    0.09
    ings
    0.09
    åIJ»
    0.09
    ful
    0.09
    aries
    0.08
    iest
    0.08
    side
    0.08
    ugin
    0.08
    orative
    0.08
    Activations Density 0.143%

    No Known Activations