© 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. 18128
    Prev
    Next
    INDEX
    Explanations

    technical code snippets and configurations

    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
     handful
    -0.09
    xp
    -0.09
    èİ·åĪ©
    -0.08
    æĴ®
    -0.08
    nier
    -0.08
    iven
    -0.08
    ource
    -0.08
    ucci
    -0.08
    大æĪ·
    -0.08
    irk
    -0.08
    POSITIVE LOGITS
    izontal
    0.16
    ancock
    0.11
    (H
    0.10
    vard
    0.10
    å·¥ä¸ļ大åѦ
    0.10
    ibur
    0.10
    ilight
    0.10
    archy
    0.10
    bard
    0.10
    archical
    0.10
    Activations Density 0.787%

    No Known Activations