© 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. 1-RES-JB
    11. 3798
    Prev
    Next
    INDEX
    Explanations

    government-related acronyms

    oai_token-act-pair · gpt-3.5-turbo

    references to governmental agencies and organizations, particularly the DOJ and HUD

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    piring
    -0.75
     Scots
    -0.73
    gradient
    -0.71
     Brunswick
    -0.69
    lihood
    -0.68
    York
    -0.64
     Camb
    -0.63
     Generations
    -0.63
    lust
    -0.62
     Columb
    -0.62
    POSITIVE LOGITS
     DOJ
    1.08
     GOODMAN
    0.87
    ODUCT
    0.79
    ILA
    0.78
    ILY
    0.73
    FBI
    0.73
    artments
    0.72
     sanctioned
    0.72
    EMENT
    0.72
     Gins
    0.71
    Activations Density 0.007%

    No Known Activations