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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 11-RESID-POST-AA
    6. 65758
    Prev
    Next
    INDEX
    Explanations

    research institutions

    np_max-act · gemini-2.0-flash

    attends to closing angle brackets and similar mark when found in the first token of a sequence.

    oai_attention-head · gpt-4oTriggered by @mlfinancebook
    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
     MVC
    -0.09
    prices
    -0.08
     plomb
    -0.08
     incline
    -0.08
     reflux
    -0.08
     merchandising
    -0.08
    坡
    -0.08
    eem
    -0.08
     ahịa
    -0.07
    -pocket
    -0.07
    POSITIVE LOGITS
    Aggreg
    0.10
    /her
    0.09
    .Aggreg
    0.09
     aggregation
    0.09
    Aggregation
    0.08
    aggreg
    0.08
     visionary
    0.08
    aggregation
    0.08
    .aggregate
    0.08
     gaz
    0.08
    Activations Density 0.018%

    No Known Activations