© 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. 57358
    Prev
    Next
    INDEX
    Explanations

    following code

    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
     Nail
    -0.09
     rather
    -0.09
    é¦Ļ
    -0.08
    éĢĶ
    -0.08
    èIJ¥éĢł
    -0.08
     gum
    -0.07
    igar
    -0.07
    ner
    -0.07
    avel
    -0.07
    éĨĴ缮
    -0.07
    POSITIVE LOGITS
    ä¹Łä¸įä¾ĭå¤ĸ
    0.10
    ç¯ij
    0.10
    襦
    0.10
    itarian
    0.10
    人çĶŁçļĦ
    0.10
    gMaps
    0.10
    entina
    0.10
    åIJ¾
    0.09
     Chá»ĭ
    0.09
    ุà¸ĩ
    0.09
    Activations Density 0.304%

    No Known Activations