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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 3-RES-JB
    11. 15875
    Prev
    Next
    INDEX
    Explanations

    mentions of geographical locations, primarily the Caribbean Islands

    oai_token-act-pair · gpt-3.5-turbo

    references to the Caribbean and its associated regions or islands

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    chen
    -0.94
    imen
    -0.93
    baugh
    -0.83
    ERAL
    -0.82
    imens
    -0.79
    WAR
    -0.78
    igun
    -0.78
    é¾įå
    -0.78
    chel
    -0.77
    eral
    -0.77
    POSITIVE LOGITS
     Caribbean
    1.10
     Islands
    1.00
     Sea
    0.91
     islands
    0.89
     Basin
    0.88
     Coast
    0.87
     Cruise
    0.87
     Indies
    0.86
     seaf
    0.86
     Seas
    0.82
    Activations Density 0.009%

    No Known Activations