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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · GPT-OSS BatchTopK SAEs
    3. GPT-OSS-20B
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 12523
    Prev
    Next
    INDEX
    Explanations

    design and presentation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-gpt-oss-20b/resid_post_layer_11/trainer_0
    Dataset (Dashboard)
    Various
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     corpus
    -0.08
     তখন
    -0.08
     separated
    -0.07
    irie
    -0.07
     fizz
    -0.07
     separados
    -0.07
    Corpus
    -0.07
     আবার
    -0.07
     indiv
    -0.07
     incont
    -0.07
    POSITIVE LOGITS
     Are
    0.08
    ווע
    0.08
     Punt
    0.08
    。我
    0.07
    ,比如
    0.07
     Airways
    0.07
     uten
    0.07
     PRO
    0.07
     bbq
    0.07
    ַ
    0.07
    Activations Density 0.137%

    No Known Activations