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. 53
    Prev
    Next
    INDEX
    Explanations

    restoration/compensation

    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
    (attrs
    -0.08
     orgas
    -0.08
    [contains
    -0.07
    'Est
    -0.07
     mall
    -0.07
     vertegenwoord
    -0.07
     amely
    -0.07
    'utiliser
    -0.07
     অন্ত
    -0.07
     Task
    -0.07
    POSITIVE LOGITS
     replen
    0.14
     compens
    0.13
     restoring
    0.12
     compensation
    0.12
     balancing
    0.12
     équilibr
    0.11
     maintaining
    0.11
     menjaga
    0.11
     compensate
    0.11
     restoration
    0.11
    Activations Density 0.011%

    No Known Activations