© 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. 57862
    Prev
    Next
    INDEX
    Explanations

    Planck institute

    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
    åģ¶
    -0.11
    çľģæĶ¿åºľ
    -0.09
    åŁºæľ¬éĿ¢
    -0.08
    åĨ
    -0.08
    çļĦèµĦéĩij
    -0.08
    á»īnh
    -0.08
    窨
    -0.08
    Drawable
    -0.08
    åħ³ç¨İ
    -0.08
    çŃ¾
    -0.08
    POSITIVE LOGITS
    erde
    0.09
    anje
    0.09
    ered
    0.09
    çŀ¬
    0.08
    éĢĶ
    0.08
    çŁ¥éģĵ
    0.08
    wash
    0.08
    èĶļ
    0.08
    年人
    0.08
    urt
    0.08
    Activations Density 0.009%

    No Known Activations