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. 21309
    Prev
    Next
    INDEX
    Explanations

    instances of the word "Gmail" or "Outlook" in the text

    oai_token-act-pair · gpt-3.5-turbo

    references to email services and associated 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
    ãĤ©
    -0.94
     Dragons
    -0.75
    dragon
    -0.73
     clad
    -0.72
    elf
    -0.71
     Strip
    -0.70
    nar
    -0.69
     displacement
    -0.68
     ice
    -0.67
     Boys
    -0.67
    POSITIVE LOGITS
     Gmail
    2.99
     Abedin
    1.97
     sender
    1.92
    Mail
    1.84
     Dropbox
    1.80
     Outlook
    1.71
     iCloud
    1.68
     AOL
    1.61
     inbox
    1.59
     Huma
    1.59
    Activations Density 0.037%

    No Known Activations