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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 8-RES-JB
    6. 244
    Prev
    Next
    INDEX
    Explanations

    mentions of the United States (U.S.) and specifically its citizens and government actions

    oai_token-act-pair · gpt-3.5-turbo

    references to the U.S. and associated numerical data

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    epad
    -0.81
    azaar
    -0.76
    ebook
    -0.75
    urger
    -0.71
    ructure
    -0.70
    oscope
    -0.68
    tch
    -0.68
    otaur
    -0.67
    atron
    -0.67
    atta
    -0.66
    POSITIVE LOGITS
     taxpayers
    1.05
     citizens
    1.00
     imperialism
    0.99
     Airways
    0.96
     diplomats
    0.96
     policymakers
    0.94
     Airlines
    0.94
     embassies
    0.93
     soil
    0.92
     Citizenship
    0.89
    Activations Density 0.105%

    No Known Activations