© 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. 16793
    Prev
    Next
    INDEX
    Explanations

    Mazda Speed or Sport

    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
    飧
    -0.09
    /licenses
    -0.09
    ries
    -0.09
    å®ģåİ¿
    -0.08
    asz
    -0.08
    ressing
    -0.08
    ÑħÑĢан
    -0.08
    asy
    -0.08
    eenth
    -0.08
    roit
    -0.08
    POSITIVE LOGITS
    檫
    0.10
    åķī
    0.10
    æ·ħ
    0.09
    çĥ¨
    0.09
    ±
    0.08
    راع
    0.08
    zone
    0.08
    hide
    0.08
    house
    0.08
    æ²¹
    0.08
    Activations Density 0.007%

    No Known Activations