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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 11189
    Prev
    Next
    INDEX
    Explanations

    mentions of specific demographic groups, particularly different ethnicities in America

    oai_token-act-pair · gpt-3.5-turbo

    references to Americans and related demographic groups

    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
     secondary
    -0.82
     detective
    -0.72
     knight
    -0.71
     external
    -0.70
     inst
    -0.69
     master
    -0.69
     option
    -0.68
     space
    -0.68
     wing
    -0.67
     Falcon
    -0.67
    POSITIVE LOGITS
    Americans
    3.63
    Democrats
    1.75
    Republicans
    1.69
    Jews
    1.68
    America
    1.55
    Austral
    1.54
    Muslims
    1.54
     Americans
    1.53
    American
    1.37
     Yorkers
    1.25
    Activations Density 0.012%

    No Known Activations