© 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. 33917
    Prev
    Next
    INDEX
    Explanations

    wanted to

    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
    äºĨèĩªå·±çļĦ
    -0.12
    ings
    -0.11
    ening
    -0.11
    ened
    -0.11
    ingly
    -0.10
    äºĨä¸Ģ个
    -0.10
    ÂŃing
    -0.10
     currently
    -0.10
    ers
    -0.10
    äºĨ好
    -0.09
    POSITIVE LOGITS
    оÑģÑĮ
    0.12
    nt
    0.11
    getic
    0.10
    ÑģÑĮ
    0.10
    apan
    0.10
    roups
    0.09
    å½ĵåĪĿ
    0.09
    æľīæĽ´å¥½çļĦ
    0.09
    ãģĦãģĨ
    0.09
    roup
    0.09
    Activations Density 0.040%

    No Known Activations