© 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. 2783
    Prev
    Next
    INDEX
    Explanations

    could not / can't + verb

    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
    æī¿æİ¥
    -0.10
    é¡·
    -0.09
    ç¿ķ
    -0.09
    ught
    -0.09
    ä½łæĥ³
    -0.09
    æݥ管
    -0.08
    åĪ¶æĪIJ
    -0.08
    毫æĹł
    -0.08
    å°±æĺ¯è¦ģ
    -0.08
    éŀį
    -0.08
    POSITIVE LOGITS
    éĵİ
    0.10
    æĦŁæħ¨
    0.10
    éĺĻ
    0.09
     Äijá»ķ
    0.09
    åĨį说
    0.09
    chas
    0.08
    asti
    0.08
    æĦŁåĬ¨
    0.08
    Extern
    0.08
    ajan
    0.08
    Activations Density 0.018%

    No Known Activations