© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50554
    Prev
    Next
    INDEX
    Explanations

    bel- prefixes

    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
    shal
    -0.10
    eer
    -0.09
    rei
    -0.09
    yard
    -0.09
    eur
    -0.09
    arak
    -0.08
    æĿ·
    -0.08
    åĪ«çļĦ
    -0.08
    à¸ĵ
    -0.08
    ÙĦÙĨ
    -0.08
    POSITIVE LOGITS
    ows
    0.11
    gium
    0.10
    ırl
    0.10
    ANCE
    0.09
    ocal
    0.09
    antly
    0.09
    oning
    0.09
    anno
    0.09
    ocities
    0.09
     unc
    0.09
    Activations Density 0.013%

    No Known Activations