© 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. 31431
    Prev
    Next
    INDEX
    Explanations

    questions and

    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
    ate
    -0.09
    æī«
    -0.09
    bet
    -0.09
    æĭħ
    -0.08
    éĹ®åį·
    -0.08
    人士
    -0.08
    壮大
    -0.08
    ÑĤом
    -0.08
    çľŁäºº
    -0.08
    æīŃæĽ²
    -0.08
    POSITIVE LOGITS
    è¿ĻäºĽéĹ®é¢ĺ
    0.10
    è¯ļæĦı
    0.09
    bearing
    0.09
    prep
    0.09
    è¡Ķ
    0.09
    åı¤ä»£
    0.09
    æŃ£å¦Ĥ
    0.09
    åºĵ
    0.09
    éļ¾é¢ĺ
    0.08
    庫
    0.08
    Activations Density 0.100%

    No Known Activations