© 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. 56038
    Prev
    Next
    INDEX
    Explanations

    views and opinions expressed

    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
    å¤įä»ĩ
    -0.10
    èĭŁ
    -0.09
    åħ¨çIJĥæľĢ大
    -0.09
    é¢Ħå¤ĩ
    -0.08
    ãģ¨ãĤĬ
    -0.08
    åĪ»æĦı
    -0.08
    prefs
    -0.08
    åŃIJåŃĻ
    -0.08
    t
    -0.08
    æľīæīĢ
    -0.08
    POSITIVE LOGITS
    ctal
    0.09
     Wo
    0.09
     Affero
    0.09
    æłĩçļĦ
    0.09
    MES
    0.09
    Responder
    0.09
    ablish
    0.08
    oader
    0.08
     tsl
    0.08
    ÑģоÑĤ
    0.08
    Activations Density 0.009%

    No Known Activations