© 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. 33918
    Prev
    Next
    INDEX
    Explanations

    McMahon

    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
    çľ¼åīį
    -0.10
    çľ¼åīįçļĦ
    -0.09
    lien
    -0.09
    ÑĥÑĩа
    -0.09
    å¤Ħå¢ĥ
    -0.08
    æľīä»·åĢ¼
    -0.08
    warts
    -0.08
     вÑĭпÑĥÑģк
    -0.08
    hood
    -0.08
    没æĶ¶
    -0.08
    POSITIVE LOGITS
    娣
    0.10
    åħħ足çļĦ
    0.09
    å¯Ŀ
    0.09
    avenport
    0.09
    è¾Ļ
    0.09
    ä¾
    0.09
    ç§ĭåĨ¬
    0.08
    太å¤ļçļĦ
    0.08
    åĭ°
    0.08
     ner
    0.08
    Activations Density 0.007%

    No Known Activations