© 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. 4043
    Prev
    Next
    INDEX
    Explanations

    Barrett, Bourne, Swanson, Smith

    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
    ovo
    -0.11
    jing
    -0.09
    tras
    -0.09
    lette
    -0.09
    äººæł¼
    -0.09
    .lv
    -0.09
    antro
    -0.08
    ä¾Ľç»Ļ侧
    -0.08
    æĢģ
    -0.08
    éĻ©
    -0.08
    POSITIVE LOGITS
    opp
    0.09
    ucci
    0.09
    ler
    0.08
    楣
    0.08
    è¿Ļ家åħ¬åı¸
    0.08
     Luk
    0.08
    ä¸įå½ĵ
    0.08
    äºĨä»ĸçļĦ
    0.08
    ç¥ļ
    0.08
    æĿ¾
    0.08
    Activations Density 0.370%

    No Known Activations