© 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. 49999
    Prev
    Next
    INDEX
    Explanations

    rough night, spot, or idea

    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
     nig
    -0.09
    /respond
    -0.09
    apro
    -0.09
    ком
    -0.09
    æĸ°å¾ģç¨ĭ
    -0.08
    泡泡
    -0.08
    QRS
    -0.08
    åĨĴ
    -0.08
    é¢łè¦Ĩ
    -0.08
    æĪijå¸ĮæľĽ
    -0.08
    POSITIVE LOGITS
    ness
    0.14
    er
    0.10
    wares
    0.10
    NESS
    0.10
    åľ°å¸¦
    0.09
    à¹Ĩ
    0.09
    ity
    0.09
    inary
    0.09
    sWith
    0.09
    ening
    0.09
    Activations Density 0.004%

    No Known Activations