© 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. 55275
    Prev
    Next
    INDEX
    Explanations

    paint

    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
    仪
    -0.11
    hood
    -0.11
    naire
    -0.10
    人éĢł
    -0.10
    å®ģæĦ¿
    -0.10
    erate
    -0.10
    ize
    -0.10
    åĦĢ
    -0.10
    ertz
    -0.09
    abet
    -0.09
    POSITIVE LOGITS
    kı
    0.09
    åľ°çIJĥ
    0.09
    LESS
    0.09
    egral
    0.09
    less
    0.08
    erv
    0.08
    lessness
    0.08
    vars
    0.08
    verter
    0.08
    castle
    0.08
    Activations Density 0.008%

    No Known Activations