© 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. 50179
    Prev
    Next
    INDEX
    Explanations

    highly active antiretroviral therapy

    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
    Haz
    -0.11
     Haz
    -0.10
    åıĪä¸Ģ次
    -0.09
    çĩ¥
    -0.09
    éĺ²ç©º
    -0.09
    ç³ľ
    -0.09
    .final
    -0.08
    æıŃ
    -0.08
    红线
    -0.08
    å¤ĩèĢĥ
    -0.08
    POSITIVE LOGITS
     Blow
    0.09
    æĭį
    0.09
    ç»ĦåIJĪ
    0.09
    BUM
    0.08
    ä¸Ģç»Ħ
    0.08
    PEG
    0.08
    æµĬ
    0.08
    롱
    0.08
     turb
    0.08
    ç¼ľ
    0.08
    Activations Density 0.073%

    No Known Activations