© 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. 61268
    Prev
    Next
    INDEX
    Explanations

    let slip, lead to, ground to, become a, set me, forecast the, road the, read the, spit take, broadcast on

    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
    di
    -0.09
     Tou
    -0.09
    kad
    -0.09
    orting
    -0.08
    anging
    -0.08
    ills
    -0.08
    lat
    -0.08
    ovation
    -0.08
    æĮŁ
    -0.08
    DIC
    -0.08
    POSITIVE LOGITS
    ings
    0.13
    able
    0.11
    äºĨä¸įå°ij
    0.11
    tees
    0.10
    äºĨä¸ĢäºĽ
    0.10
    äºĨä¸Ģ个
    0.10
    ances
    0.09
    äºĨèĩªå·±çļĦ
    0.09
    äºĨå¾Īå¤ļ
    0.09
    swith
    0.09
    Activations Density 0.119%

    No Known Activations