© 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. 1-RES-JB
    11. 3099
    Prev
    Next
    INDEX
    Explanations

    references to countries, specifically Morocco

    oai_token-act-pair · gpt-3.5-turbo

    references to Morocco and Algeria

    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
    ellation
    -0.85
    rophe
    -0.82
    sworth
    -0.79
    eve
    -0.79
    ritch
    -0.75
    ext
    -0.72
    icle
    -0.72
    ttle
    -0.70
    riet
    -0.70
    weeney
    -0.70
    POSITIVE LOGITS
     Morocco
    1.16
     Algeria
    0.98
     Alger
    0.93
     Sahara
    0.91
     Moroccan
    0.90
     Tunisia
    0.84
     Arabia
    0.83
     Ré
    0.78
    CLASSIFIED
    0.78
     Arabian
    0.78
    Activations Density 0.013%

    No Known Activations