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. 19205
    Prev
    Next
    INDEX
    Explanations

    Numbers found in equations

    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
    _compare
    -0.08
     тез
    -0.08
     иден
    -0.08
    ighbours
    -0.08
     Отлич
    -0.08
    uldade
    -0.07
    _enable
    -0.07
     увидеть
    -0.07
    ities
    -0.07
    istrik
    -0.07
    POSITIVE LOGITS
     admittedly
    0.09
    हाल
    0.08
     actually
    0.08
    —or
    0.08
     repos
    0.08
    ín
    0.08
     artículos
    0.07
     apat
    0.07
    yage
    0.07
    ِي
    0.07
    Activations Density 0.068%

    No Known Activations