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. 131007
    Prev
    Next
    INDEX
    Explanations

    say "instruction"

    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
    芽
    -0.09
     Clube
    -0.08
    еди
    -0.08
     пики
    -0.08
     bayi
    -0.08
     koup
    -0.08
    .sqrt
    -0.08
     mutants
    -0.08
    Afee
    -0.08
    emers
    -0.08
    POSITIVE LOGITS
     instruction
    0.57
    _instruction
    0.53
    Instruction
    0.52
    instruction
    0.52
     Instruction
    0.52
     instruk
    0.45
     инструкция
    0.44
     инструк
    0.44
     instructions
    0.42
     instrucciones
    0.42
    Activations Density 0.044%

    No Known Activations