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. 23-RESID-POST-AA
    6. 130985
    Prev
    Next
    INDEX
    Explanations

    numbers (tens)

    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_23/trainer_0
    Dataset (Dashboard)
    Various
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Let's
    -0.12
     That's
    -0.10
     Actually
    -0.10
     Maybe
    -0.10
                                                                   
    -0.09
     For
    -0.09
    ġġ
    -0.09
     �
    -0.09
     ***↵
    -0.09
    illuni
    -0.09
    POSITIVE LOGITS
    (\'
    0.11
     twenty
    0.10
     \"$
    0.10
     thirty
    0.10
    	the
    0.10
     eighty
    0.10
     forty
    0.10
     mankind
    0.10
     \'
    0.10
     thirteen
    0.10
    Activations Density 3.399%

    No Known Activations