© 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. 33689
    Prev
    Next
    INDEX
    Explanations

    but followed by contrastive conjunctions

    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
    èĻ¹æ¡¥
    -0.09
    endregion
    -0.09
     neglig
    -0.09
    è¿Ļåıª
    -0.09
    ERO
    -0.09
    å°¤åħ¶æĺ¯
    -0.09
    å°¤åħ¶
    -0.08
    åIJİèĢħ
    -0.08
    fil
    -0.08
    étr
    -0.08
    POSITIVE LOGITS
    /of
    0.14
    éĥ½æľīèĩªå·±
    0.12
    /or
    0.11
    æĿµ
    0.10
    ä¼ļä¼ļéķ¿
    0.10
    asca
    0.09
    ients
    0.09
    大大å°ıå°ı
    0.09
    ä½Ĩä¸įéĻIJäºİ
    0.09
     Fowler
    0.08
    Activations Density 0.088%

    No Known Activations