© 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. 39627
    Prev
    Next
    INDEX
    Explanations

    humor and satire

    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
    ä»ĭç»įä¸Ģä¸ĭ
    -0.09
    俦
    -0.08
     instinct
    -0.08
    æĬ¤
    -0.08
    .blur
    -0.08
    دÙĪ
    -0.08
    éĵ¤
    -0.08
    ä¸įçĪ±
    -0.08
    åıijè¡Į
    -0.08
     CHANGE
    -0.08
    POSITIVE LOGITS
    ç¬ijè¯Ŀ
    0.21
    åĺ²ç¬ij
    0.20
    ç¬ijçĿĢ
    0.18
    è°ĥä¾ĥ
    0.17
    è¯Ļ
    0.17
    ç¬ijäºĨ
    0.17
    åĺ²
    0.17
    ç¬ij
    0.17
    讥
    0.17
     laugh
    0.16
    Activations Density 0.061%

    No Known Activations