© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 11-RES-JB
    6. 900
    Prev
    Next
    INDEX
    Explanations

    historical figures and events related to English royalty

    oai_token-act-pair · gpt-3.5-turbo

    historical references to English monarchy and royal figures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    kaya
    -1.00
    nikov
    -0.91
    imov
    -0.88
     DOE
    -0.87
    achev
    -0.83
    taboola
    -0.83
     Yuri
    -0.82
    SCP
    -0.82
    externalActionCode
    -0.82
     zipper
    -0.81
    POSITIVE LOGITS
     Cornwall
    1.45
     Scots
    1.45
     Suffolk
    1.40
     Glou
    1.39
     Welsh
    1.38
     Britons
    1.36
     Scotland
    1.34
     Wales
    1.31
     Ulster
    1.31
     Somerset
    1.30
    Activations Density 0.316%

    No Known Activations