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. 11-RESID-POST-AA
    6. 88277
    Prev
    Next
    INDEX
    Explanations

    finance and investment

    np_max-act · gemini-2.0-flash

    attends from the beginning of a sequence to any sequence-initial tokens appearing earlier.

    oai_attention-head · gpt-4oTriggered by @mlfinancebook

    the neuron is looking for information about share certificates and ownership documentation.

    oai_token-act-pair · gpt-5-nanoTriggered by @tatsatx
    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
    _cli
    -0.08
     itself
    -0.07
    ült
    -0.07
    arsim
    -0.07
    検索
    -0.07
    -cli
    -0.07
    <context
    -0.07
     cli
    -0.07
     напис
    -0.07
    ленно
    -0.07
    POSITIVE LOGITS
    收益
    0.12
     dividends
    0.11
    Dividend
    0.10
    赚
    0.10
     keuntungan
    0.10
     revenues
    0.10
     profits
    0.10
    Income
    0.09
     revenus
    0.09
     dividend
    0.09
    Activations Density 0.012%

    No Known Activations