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. 16218
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "marriage," particularly in the context of important or controversial events

    oai_token-act-pair · gpt-3.5-turbo

    references to marriage, particularly focusing on same-sex marriage

    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
     istg
    -0.71
    ACC
    -0.68
     Sho
    -0.66
    ops
    -0.66
     Zap
    -0.65
    owl
    -0.64
    wo
    -0.64
    ensor
    -0.63
    ateurs
    -0.63
    alert
    -0.62
    POSITIVE LOGITS
     marriage
    3.73
     Marriage
    3.04
    marriage
    2.89
     marriages
    2.88
     marrying
    2.01
     marry
    1.97
     wedding
    1.94
     marital
    1.86
     divorce
    1.85
     married
    1.85
    Activations Density 0.017%

    No Known Activations