Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 17742
    Prev
    Next
    INDEX
    Explanations

    user-related keywords, specifically focusing on an individual's name or login identity

    oai_token-act-pair · gpt-3.5-turbo

    references to user identifiers or user-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
     funer
    -0.79
     knots
    -0.77
     Bapt
    -0.67
     Estimates
    -0.66
     beads
    -0.66
     weddings
    -0.65
     mourning
    -0.65
     ox
    -0.65
     embassies
    -0.64
     governors
    -0.63
    POSITIVE LOGITS
    user
    4.17
    users
    2.89
    USER
    2.30
    User
    2.13
     user
    1.82
     User
    1.74
    use
    1.54
    username
    1.53
     Users
    1.50
    Users
    1.46
    Activations Density 0.014%

    No Known Activations