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. 9049
    Prev
    Next
    INDEX
    Explanations

    instances of exchanging items or actions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "exchanged" and related terms

    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
     ADHD
    -0.72
     Project
    -0.64
    Neil
    -0.62
     Trey
    -0.62
     sci
    -0.61
    aly
    -0.60
     Nightmares
    -0.60
     tru
    -0.59
    rey
    -0.59
    fig
    -0.59
    POSITIVE LOGITS
     exchanged
    3.85
     exchanging
    2.38
     exchange
    2.03
     exchanges
    1.95
     swapped
    1.87
     traded
    1.67
     Exchange
    1.32
     parted
    1.31
     exch
    1.29
     interchange
    1.23
    Activations Density 0.013%

    No Known Activations