© 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. 4-RES-JB
    11. 874
    Prev
    Next
    INDEX
    Explanations

    mentions of Latin American and Central American countries

    oai_token-act-pair · gpt-3.5-turbo

    references to geographical and cultural identities, particularly related to American, African, and Asian contexts

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

    No Comments

    Negative Logits
    paio
    -0.87
    wright
    -0.79
    tyard
    -0.74
    CAP
    -0.73
    rex
    -0.73
     Warner
    -0.73
    nc
    -0.73
    tesy
    -0.72
    trak
    -0.70
     Charlottesville
    -0.69
    POSITIVE LOGITS
     aspir
    0.93
     captives
    0.84
     civilisation
    0.82
     fri
    0.82
     diplomats
    0.82
     nationals
    0.81
     replacements
    0.79
     descent
    0.79
     exile
    0.78
     equivalents
    0.77
    Activations Density 0.054%

    No Known Activations