© 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. 34816
    Prev
    Next
    INDEX
    Explanations

    que

    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
    nodoc
    -0.09
    åĽĽé¡¹
    -0.09
    æĻļä¼ļ
    -0.09
     мил
    -0.09
    æīĵäºĨ
    -0.09
    æĻļéĹ´
    -0.09
    çĶ¥
    -0.09
    æĭįæĭį
    -0.09
    è´£ç¼ĸ
    -0.09
     nghá»ģ
    -0.09
    POSITIVE LOGITS
    br
    0.09
    brane
    0.08
    å²Ĺ
    0.08
    etting
    0.08
    ijo
    0.08
    ÑĢÑĥ
    0.08
    åĪļ
    0.08
    pid
    0.08
    enary
    0.07
    åīĽ
    0.07
    Activations Density 0.074%

    No Known Activations