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

    Neuronpedia

    Jacobian LensNEW
    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. 2-RES-JB
    11. 516
    Prev
    Next
    INDEX
    Explanations

    titles of royalty, specifically the word "Emperor"

    oai_token-act-pair · gpt-4-turbo

    references to historical figures, specifically emperors

    oai_token-act-pair · gpt-3.5-turbo

    references to emperors and imperial figures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ter
    -0.79
    chel
    -0.77
    trak
    -0.73
    asket
    -0.73
    esville
    -0.72
    ties
    -0.71
    tera
    -0.71
    nel
    -0.70
    crew
    -0.69
    dos
    -0.67
    POSITIVE LOGITS
    peror
    1.07
     Nero
    1.04
    perors
    1.02
     emperor
    0.92
     Emperor
    0.92
     pengu
    0.90
    esses
    0.83
     Napoleon
    0.81
     Claud
    0.79
     Augustus
    0.75
    Activations Density 0.008%

    No Known Activations