© 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. 54841
    Prev
    Next
    INDEX
    Explanations

    intense emotional expression

    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¬Ģ
    -0.11
    è°ĭ
    -0.10
    plode
    -0.09
    ÐľÐŀ
    -0.09
    linger
    -0.09
    æIJŃ
    -0.09
    å¨ģæħij
    -0.09
    å§ĵ
    -0.09
    ä½ľåĩºäºĨ
    -0.09
    æ¯ĶåĪ©
    -0.09
    POSITIVE LOGITS
    æ¿ĢçĥĪçļĦ
    0.13
     intensity
    0.10
    è¯Ńæ°Ķ
    0.10
     intense
    0.09
    ä¸ĢæŶ
    0.09
    æ¿ĢçĥĪ
    0.09
    ly
    0.09
     passion
    0.08
    Intensity
    0.08
    åĬłç´§
    0.08
    Activations Density 0.053%

    No Known Activations