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. 59525
    Prev
    Next
    INDEX
    Explanations

    Calculating totals

    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
     lata
    -0.08
     narrowly
    -0.08
     мало
    -0.08
     свеж
    -0.08
    emails
    -0.08
     encaps
    -0.08
    Св
    -0.08
     Asoci
    -0.08
     ಕಡ
    -0.08
    숨
    -0.07
    POSITIVE LOGITS
     becomes
    0.09
     يصبح
    0.09
     parity
    0.08
    total
    0.08
     divent
    0.08
    صبح
    0.08
    ,total
    0.08
    Tak
    0.08
    Total
    0.08
     devient
    0.08
    Activations Density 0.102%

    No Known Activations