© 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. 10-RES-JB
    11. 20243
    Prev
    Next
    INDEX
    Explanations

    references to United States history and politics

    oai_token-act-pair · gpt-3.5-turbo

    references to American history or culture

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     fielder
    -0.80
    ouf
    -0.77
     Dalai
    -0.70
     inhibitor
    -0.70
    uilt
    -0.69
    viks
    -0.68
    rogens
    -0.64
     Duchess
    -0.63
    Hour
    -0.62
     Mermaid
    -0.62
    POSITIVE LOGITS
     society
    1.33
     politics
    1.33
     history
    1.19
     affairs
    1.09
     capitalism
    1.08
     folklore
    1.08
     circles
    1.08
     academia
    1.07
     culture
    1.06
     civilization
    1.03
    Activations Density 0.269%

    No Known Activations