© 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. 27308
    Prev
    Next
    INDEX
    Explanations

    adapted from

    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
    untu
    -0.11
    è¿ĩåī©
    -0.10
    á»ĥn
    -0.10
    odi
    -0.10
    è¿ĩ度
    -0.09
    åıijçİ°éĹ®é¢ĺ
    -0.09
    èĦĨ
    -0.09
    æĸĭ
    -0.09
    çĶ·ä¸»è§Ĵ
    -0.09
    强æĤį
    -0.09
    POSITIVE LOGITS
    æĿ¥æºIJäºİ
    0.11
    æºIJäºİ
    0.10
    มาà¸Īาà¸ģ
    0.10
    çģµæĦŁ
    0.09
    avad
    0.09
    æijĺ
    0.09
     excerpt
    0.09
     Registry
    0.09
     based
    0.09
    Registry
    0.09
    Activations Density 0.149%

    No Known Activations