© 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. 23392
    Prev
    Next
    INDEX
    Explanations

    administered by

    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
    ifying
    -0.11
     зÑĢениÑı
    -0.10
    楮
    -0.10
    tern
    -0.10
    /from
    -0.10
    vant
    -0.09
    ific
    -0.09
    ãĢħ
    -0.09
    ifier
    -0.09
    еÑĩ
    -0.09
    POSITIVE LOGITS
    istrator
    0.14
    istration
    0.13
    ãģĹãģ¾ãģĨ
    0.11
    stration
    0.11
    ISTR
    0.11
    ance
    0.10
    lte
    0.10
    anger
    0.10
    eer
    0.10
    lı
    0.09
    Activations Density 0.044%

    No Known Activations