© 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. 61403
    Prev
    Next
    INDEX
    Explanations

    "the" followed by a noun phrase

    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
    线ä¸Ĭ线ä¸ĭ
    -0.10
    ä¸Ĭ线
    -0.09
    OnClick
    -0.09
    ows
    -0.09
    ï¸ı
    -0.09
    èĸ®
    -0.09
    åľ¨æľªæĿ¥
    -0.09
    ä½łåľ¨
    -0.09
     onsite
    -0.09
    æĻ°
    -0.08
    POSITIVE LOGITS
     behalf
    0.30
    /off
    0.17
     occasion
    0.13
     verge
    0.12
    side
    0.12
     оÑģнове
    0.12
    slaught
    0.11
    обоÑĢоÑĤ
    0.11
    çħ
    0.11
     eve
    0.11
    Activations Density 0.116%

    No Known Activations