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. 11-RES-JB
    6. 10946
    Prev
    Next
    INDEX
    Explanations

    mentions of actions related to the U.S. National Anthem, particularly in a controversial context

    oai_token-act-pair · gpt-3.5-turbo

    references to performance of nationalistic acts or ceremonies by public figures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Lisbon
    -0.95
     EF
    -0.85
     Macro
    -0.82
     Bord
    -0.81
     Dracula
    -0.80
     Tud
    -0.79
     Mirage
    -0.79
     cellar
    -0.75
    oming
    -0.75
     Belg
    -0.73
    POSITIVE LOGITS
     Kaepernick
    2.34
    aepernick
    2.10
     kneeling
    1.77
     anthem
    1.62
     kne
    1.50
    bleacher
    1.41
    NFL
    1.34
     protesting
    1.23
     NFL
    1.23
     Colin
    1.22
    Activations Density 0.212%

    No Known Activations