© 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. 50346
    Prev
    Next
    INDEX
    Explanations

    snowboarding and snow sports

    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
    nea
    -0.11
    ../../../
    -0.10
    suffix
    -0.10
    æĥ®
    -0.10
    åİ¢
    -0.10
    à¸ŀล
    -0.10
    룬
    -0.10
    éĶ¦ç»£
    -0.09
    RARY
    -0.09
    suming
    -0.09
    POSITIVE LOGITS
    flake
    0.12
    ed
    0.09
    boarding
    0.09
    è¡·
    0.09
    -haired
    0.08
    åIJĦå¼ı
    0.08
    est
    0.08
    lö
    0.08
    caled
    0.08
    vale
    0.08
    Activations Density 0.008%

    No Known Activations