© 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. 5914
    Prev
    Next
    INDEX
    Explanations

    summing or sum

    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
    常å§Ķ
    -0.12
    ector
    -0.10
    ye
    -0.10
    gow
    -0.10
    ampo
    -0.09
    aven
    -0.09
     Amir
    -0.09
    ãģĬãģı
    -0.09
    zell
    -0.09
    enen
    -0.08
    POSITIVE LOGITS
    MARY
    0.17
    maries
    0.15
    enance
    0.14
    mand
    0.13
    mers
    0.13
    atoria
    0.13
    ming
    0.12
    åķľ
    0.11
    ative
    0.11
    mere
    0.10
    Activations Density 0.067%

    No Known Activations