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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 5526
    Prev
    Next
    INDEX
    Explanations

    the word "Gerry" at different activation levels

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the word "gerry" or variations of it, typically in the context of political gerrymandering

    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
    Trend
    -0.73
    href
    -0.69
    resses
    -0.68
    urg
    -0.67
    eki
    -0.66
    viol
    -0.64
    escal
    -0.64
    ologists
    -0.64
     heartbeat
    -0.63
    agos
    -0.63
    POSITIVE LOGITS
    erry
    1.39
     Gerry
    0.90
    mand
    0.85
    mite
    0.82
    jee
    0.82
    mallow
    0.75
    wu
    0.75
    weed
    0.74
    bye
    0.73
    leigh
    0.73
    Activations Density 0.006%

    No Known Activations