© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 8-RES-JB
    11. 1797
    Prev
    Next
    INDEX
    Explanations

    major entities or key players in various contexts, such as countries, companies, or organizations

    oai_token-act-pair · gpt-3.5-turbo

    references to major countries or entities involved in global issues

    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
     Reincarn
    -0.81
    potion
    -0.77
    eat
    -0.71
    urate
    -0.71
    hiro
    -0.70
    mite
    -0.70
     Rampage
    -0.69
    gery
    -0.66
    nothing
    -0.66
    atcher
    -0.66
    POSITIVE LOGITS
     countries
    1.41
     institutions
    1.40
     nations
    1.38
     sectors
    1.29
     firms
    1.28
     industries
    1.27
     agencies
    1.24
     categories
    1.23
     companies
    1.23
     organisations
    1.23
    Activations Density 0.534%

    No Known Activations