© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 8-RES-JB
    11. 13138
    Prev
    Next
    INDEX
    Explanations

    instances of texts indicating a financial transaction or exchange

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating payment or financial transactions

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    soType
    -0.80
    aiman
    -0.76
    ilda
    -0.75
    gdala
    -0.72
    ItemThumbnailImage
    -0.71
    enth
    -0.70
    aha
    -0.70
    erning
    -0.69
    itta
    -0.69
    hew
    -0.68
    POSITIVE LOGITS
     upkeep
    1.04
     necessities
    0.95
     groceries
    0.92
     extras
    0.90
     postage
    0.89
     services
    0.85
     tickets
    0.85
     subscriptions
    0.84
     transportation
    0.82
     repairs
    0.80
    Activations Density 0.078%

    No Known Activations