Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 21343
    Prev
    Next
    INDEX
    Explanations

    words related to geographical regions, particularly the Middle East

    oai_token-act-pair · gpt-3.5-turbo

    references to the Middle East

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     disg
    -0.68
    crew
    -0.67
     gag
    -0.66
    rics
    -0.64
    burst
    -0.64
    Vo
    -0.64
    orpor
    -0.62
    ickr
    -0.62
    rays
    -0.62
    proof
    -0.62
    POSITIVE LOGITS
     Middle
    3.87
    Middle
    3.10
    middle
    1.90
     middle
    1.70
     Arab
    1.54
     Arabian
    1.45
     MID
    1.39
     Balkans
    1.34
    Arab
    1.33
     Persian
    1.32
    Activations Density 0.015%

    No Known Activations