© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52377
    Prev
    Next
    INDEX
    Explanations

    witty, sassy, sarcastic

    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
    è¯ļæĦı
    -0.11
    angler
    -0.10
    å®°
    -0.09
    ç´Ķ
    -0.09
    æĹłå£°
    -0.09
    饮
    -0.09
     Awakening
    -0.09
    å¿ĥæĦ¿
    -0.08
     choking
    -0.08
    æ¶
    -0.08
    POSITIVE LOGITS
    lien
    0.10
    è°ĥä¾ĥ
    0.10
    tical
    0.09
     IDC
    0.09
     observ
    0.09
    ä½»
    0.08
     sharp
    0.08
     witty
    0.08
    tics
    0.08
    带éĺŁ
    0.08
    Activations Density 0.036%

    No Known Activations