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. 15-RESID-POST-AA
    6. 18909
    Prev
    Next
    INDEX
    Explanations

    40

    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
     erhe
    -0.09
     huge
    -0.09
     придется
    -0.08
     상당
    -0.08
     numerous
    -0.08
     möglicherweise
    -0.08
     énorme
    -0.08
     hätte
    -0.07
     Various
    -0.07
     çeşitli
    -0.07
    POSITIVE LOGITS
     optimal
    0.15
    Optimal
    0.15
     optimum
    0.13
     Optimal
    0.13
    optimal
    0.13
     preferably
    0.13
     desirable
    0.12
     optimaal
    0.12
     optimale
    0.12
     ideally
    0.12
    Activations Density 0.026%

    No Known Activations