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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 1-RES-JB
    11. 14210
    Prev
    Next
    INDEX
    Explanations

    references to specific countries, with a focus on Ghana and Zambia

    oai_token-act-pair · gpt-3.5-turbo

    mentions of specific African countries

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    sed
    -0.77
    ellation
    -0.77
    chel
    -0.75
    colo
    -0.74
    FOX
    -0.72
    essler
    -0.70
    cone
    -0.70
    nda
    -0.70
    ueller
    -0.69
    croft
    -0.69
    POSITIVE LOGITS
     Tanzania
    1.25
     Ghana
    1.17
     Cameroon
    1.13
     Senegal
    1.08
     Congo
    1.07
     Nigeria
    1.03
     Zimbabwe
    1.01
     Guinea
    1.01
     Angola
    0.99
     Uganda
    0.98
    Activations Density 0.016%

    No Known Activations