© 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. 18256
    Prev
    Next
    INDEX
    Explanations

    instances of the term "PAT" followed by a number

    oai_token-act-pair · gpt-3.5-turbo

    references to the PATRIOT Act

    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
    zig
    -0.74
    packing
    -0.70
    comments
    -0.70
    terness
    -0.68
    reth
    -0.67
    rage
    -0.66
    ?]
    -0.66
    VERSION
    -0.66
    guy
    -0.66
    ensation
    -0.64
    POSITIVE LOGITS
    DragonMagazine
    0.77
     berth
    0.77
     behavi
    0.69
     cavity
    0.66
     compartment
    0.63
     jammed
    0.63
     PAT
    0.62
     cylinder
    0.62
     electr
    0.61
     liberty
    0.61
    Activations Density 0.001%

    No Known Activations