© 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. 52846
    Prev
    Next
    INDEX
    Explanations

    who or that followed by a verb

    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
    æĭĹ
    -0.09
    à°¬
    -0.08
    ming
    -0.08
    åĻĤ
    -0.08
    çīŁ
    -0.07
    å¿ħè¦ģ
    -0.07
    æĻĥ
    -0.07
    大å¤ļ
    -0.07
    ¤
    -0.07
    yro
    -0.07
    POSITIVE LOGITS
    èĥ½æĬĬ
    0.13
    éĤ£ä¸ªäºº
    0.11
    æĩĤå¾Ĺ
    0.11
    èĥ½
    0.11
    èĥ½å¤Ł
    0.11
    æľīä¸įåIJĮçļĦ
    0.11
    æĹ¢èĥ½
    0.11
    æľīèĥ½åĬĽ
    0.10
    èĥ½å¤ł
    0.10
    æľīä¿¡å¿ĥ
    0.10
    Activations Density 0.096%

    No Known Activations