Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · GPT-OSS BatchTopK SAEs
    3. GPT-OSS-20B
    4. Resid Post - 131k
    5. 15-RESID-POST-AA
    6. 60
    Prev
    Next
    INDEX
    Explanations

    Combining or simplifying

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-gpt-oss-20b/resid_post_layer_15/trainer_0
    Dataset (Dashboard)
    Various
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     many
    -0.09
     systemic
    -0.09
     العديد
    -0.08
    Many
    -0.08
     otherwise
    -0.08
    Expired
    -0.08
     muitos
    -0.08
    :selected
    -0.08
    .health
    -0.08
    .on
    -0.08
    POSITIVE LOGITS
     simplifies
    0.10
    组合
    0.10
     Combining
    0.10
     Combination
    0.10
     simplify
    0.09
     Simpl
    0.09
     combinación
    0.09
    简
    0.09
     संय
    0.09
     combina
    0.09
    Activations Density 0.022%

    No Known Activations