© 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. 56950
    Prev
    Next
    INDEX
    Explanations

    joking or kidding

    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
    istar
    -0.11
    è¹Ī
    -0.10
    antas
    -0.10
     STAR
    -0.10
    ioni
    -0.10
     conspic
    -0.10
    çıij
    -0.09
    è¡°
    -0.09
    acic
    -0.09
    /star
    -0.09
    POSITIVE LOGITS
    çİ©ç¬ij
    0.20
    å¼Ģçİ©ç¬ij
    0.18
    å¹½é»ĺ
    0.15
    æĪı
    0.15
    æĪ²
    0.12
     joking
    0.12
    è°ĥä¾ĥ
    0.12
    å¬ī
    0.12
     kidding
    0.12
     joked
    0.11
    Activations Density 0.072%

    No Known Activations