© 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. 54519
    Prev
    Next
    INDEX
    Explanations

    Steele or steel names

    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
    inals
    -0.09
    quot
    -0.08
    ponents
    -0.08
    stead
    -0.08
    Ñģко
    -0.08
    妥
    -0.08
    nung
    -0.08
    群
    -0.08
    èĤļ
    -0.08
    åŃĹåı·
    -0.07
    POSITIVE LOGITS
    ently
    0.11
    à¹Ģà¸Ńา
    0.10
    lish
    0.09
    /cs
    0.09
    argest
    0.08
    etr
    0.08
    æijĨ
    0.08
    hift
    0.08
    LES
    0.08
    æĪij以为
    0.08
    Activations Density 0.011%

    No Known Activations