© 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. 1-RES-JB
    11. 21605
    Prev
    Next
    INDEX
    Explanations

    references to specific political figures, particularly Chuck Schumer

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the politician Chuck Schumer

    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
    resp
    -0.86
    rial
    -0.81
     Virgin
    -0.73
     Georgian
    -0.72
     Columb
    -0.70
    LEASE
    -0.67
    held
    -0.64
    lict
    -0.64
    WARD
    -0.62
     Gutenberg
    -0.61
    POSITIVE LOGITS
     Schumer
    1.33
     shenan
    0.90
    ikuman
    0.82
     Tanz
    0.74
    hler
    0.73
    steen
    0.72
    arella
    0.72
    atown
    0.71
    idy
    0.71
     Grassley
    0.71
    Activations Density 0.006%

    No Known Activations