© 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. 22929
    Prev
    Next
    INDEX
    Explanations

    Lehman, Umrah, Gleason

    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
    æīŀ
    -0.10
    Õļ
    -0.10
    inson
    -0.10
    åıĬ缸åħ³
    -0.10
    terior
    -0.09
    wu
    -0.09
    ã¨
    -0.09
    å¹³çŃī
    -0.09
    æĸ°é²ľ
    -0.09
    æݧèĤ¡éĽĨåĽ¢
    -0.09
    POSITIVE LOGITS
    åħį
    0.08
    Âł
    0.08
    ATAB
    0.08
    åĩºåħ·
    0.07
    0
    0.07
    éĢĤ度
    0.07
    5
    0.07
    æ±Ģ
    0.07
    IT
    0.07
    rus
    0.07
    Activations Density 0.485%

    No Known Activations