© 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. 3-RES-JB
    11. 10709
    Prev
    Next
    INDEX
    Explanations

    words related to the United Arab Emirates

    oai_token-act-pair · gpt-3.5-turbo

    references to the United Arab Emirates (UAE) and its related contexts

    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
    istic
    -0.84
    istically
    -0.84
    fig
    -0.80
    avery
    -0.79
    ests
    -0.74
    istics
    -0.74
    aster
    -0.74
    izing
    -0.71
    auld
    -0.70
    izations
    -0.67
    POSITIVE LOGITS
     Emir
    1.11
     Emirates
    1.05
     Dhabi
    1.02
     UAE
    0.93
    rahim
    0.85
     Cooperation
    0.84
    uran
    0.76
     Arabian
    0.74
     Airlines
    0.74
    doms
    0.72
    Activations Density 0.009%

    No Known Activations