© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 40387
    Prev
    Next
    INDEX
    Explanations

    fled or flee

    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
    Negative Logits
    ually
    -0.11
    ity
    -0.10
     ÑģебÑı
    -0.10
    hift
    -0.09
    è¡®
    -0.09
    /people
    -0.09
    ëıĮ
    -0.09
    éĻ²
    -0.08
    aceous
    -0.08
    ally
    -0.08
    POSITIVE LOGITS
    chner
    0.10
    otec
    0.10
    èº
    0.09
    nder
    0.09
    gable
    0.09
    ÑģÑĤвие
    0.09
    桧
    0.09
    ande
    0.09
    داد
    0.09
    uren
    0.09
    Activations Density 0.005%

    No Known Activations