© 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. 4-RES-JB
    11. 4899
    Prev
    Next
    INDEX
    Explanations

    mentions of specific names, particularly "Nikki Haley" and "Daniels", potentially related to politics

    oai_token-act-pair · gpt-3.5-turbo

    references to specific political figures, particularly Nikki Haley

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

    No Comments

    Negative Logits
    orate
    -0.76
    ~~~~~~~~~~~~~~~~
    -0.76
    Else
    -0.75
    book
    -0.71
    ically
    -0.70
    ISTER
    -0.69
    Copyright
    -0.68
    amins
    -0.67
    izarre
    -0.67
    abol
    -0.65
    POSITIVE LOGITS
     Haley
    1.02
    burg
    0.95
    burgh
    0.79
    sung
    0.68
     peppers
    0.68
    assetsadobe
    0.67
    uli
    0.66
    endas
    0.66
    bean
    0.66
    otte
    0.65
    Activations Density 0.009%

    No Known Activations