© 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. 11-RES-JB
    11. 23159
    Prev
    Next
    INDEX
    Explanations

    the name "Angela Merkel."

    oai_token-act-pair · gpt-3.5-turbo

    references to Angela Merkel, particularly in the context of her role as Chancellor of Germany

    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
    amins
    -0.77
    omon
    -0.77
     skelet
    -0.67
    inqu
    -0.66
    ycle
    -0.66
    sole
    -0.66
    essee
    -0.63
    omin
    -0.62
    bish
    -0.62
    ask
    -0.61
    POSITIVE LOGITS
     Merkel
    1.59
    llor
    0.93
     Duck
    0.84
     Lans
    0.82
     Hollande
    0.75
     Hendricks
    0.74
     Angela
    0.74
     Chancellor
    0.74
     Eagle
    0.73
     Thatcher
    0.72
    Activations Density 0.028%

    No Known Activations