© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 10318
    Prev
    Next
    INDEX
    Explanations

    references to software packages

    oai_token-act-pair · gpt-3.5-turbo

    mentions of packages or package-related concepts

    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
    姫
    -0.70
     friendships
    -0.68
     Kenobi
    -0.67
     ruled
    -0.66
     Idle
    -0.66
    zig
    -0.65
    天
    -0.65
    bec
    -0.65
     Draper
    -0.64
    atell
    -0.64
    POSITIVE LOGITS
     package
    3.82
     packages
    2.92
     Package
    2.67
    package
    2.52
    Package
    2.17
    packages
    1.93
     bundle
    1.73
     packaging
    1.67
     PACK
    1.61
     packet
    1.50
    Activations Density 0.021%

    No Known Activations