© 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. 5225
    Prev
    Next
    INDEX
    Explanations

    job that make our determined back message

    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
     dá»±ng
    -0.08
    _rg
    -0.08
    åįĹçľģ
    -0.08
    RV
    -0.08
    RY
    -0.08
    á»ĩm
    -0.08
    éĵĸ
    -0.08
    èł²
    -0.08
    ä¸ªåĽ½å®¶
    -0.08
    çĪ¹
    -0.08
    POSITIVE LOGITS
    عت
    0.09
    lobby
    0.08
    ä¸Ģå¿ĥ
    0.08
    .await
    0.08
     Pant
    0.08
    èıľ
    0.08
    对èĩªå·±çļĦ
    0.08
    æ½ľ
    0.08
    客æĪ¿
    0.08
    iras
    0.08
    Activations Density 0.676%

    No Known Activations