© 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. 26441
    Prev
    Next
    INDEX
    Explanations

    logging in or signs in

    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
    稷
    -0.10
    åľ¨æł¡
    -0.10
    tim
    -0.09
    åħ¶
    -0.09
    fully
    -0.09
    mente
    -0.09
    dens
    -0.09
    åĪĹåħ¥
    -0.08
    antor
    -0.08
    ÐIJÐŀ
    -0.08
    POSITIVE LOGITS
    serter
    0.13
    /out
    0.12
    otify
    0.10
    fty
    0.10
    verted
    0.10
    lays
    0.09
    æİĩ
    0.09
    ferences
    0.08
    shore
    0.08
    fragistics
    0.08
    Activations Density 0.062%

    No Known Activations