© 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. 35269
    Prev
    Next
    INDEX
    Explanations

    colorless, pale, metal

    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
    sted
    -0.10
    pson
    -0.09
    ering
    -0.09
    没æĶ¶
    -0.09
    èĢį
    -0.08
    æ²Ĩ
    -0.08
    æī«
    -0.08
    åī©ä¸ĭçļĦ
    -0.08
    æļ´
    -0.08
    è¿Ļ份
    -0.08
    POSITIVE LOGITS
    çĻ½èī²çļĦ
    0.11
    é»Ħèī²
    0.11
    çĬ¶
    0.11
    çĻ½èī²
    0.11
    çĭĢ
    0.10
    çľĭä¸įè§ģ
    0.10
     appearance
    0.10
    éĢıæĺİ
    0.10
    redd
    0.09
    èĭįçĻ½
    0.09
    Activations Density 0.142%

    No Known Activations