© 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. 2039
    Prev
    Next
    INDEX
    Explanations

    class declaration a

    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
    nya
    -0.10
    ta
    -0.10
    sw
    -0.10
    ت
    -0.10
    t
    -0.09
    avit
    -0.09
    vv
    -0.09
    lies
    -0.09
    aries
    -0.09
    gun
    -0.09
    POSITIVE LOGITS
    egis
    0.13
    Ùĭ
    0.13
     aVar
    0.12
    uteur
    0.11
    ffect
    0.11
    ught
    0.11
    arhus
    0.11
    ustralian
    0.11
    alto
    0.11
    ffects
    0.10
    Activations Density 0.283%

    No Known Activations