Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 7718
    Prev
    Next
    INDEX
    Explanations

    references to government spending or budget allocations

    oai_token-act-pair · gpt-3.5-turbo

    references to government spending

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Sie
    -0.74
    ĪĴ
    -0.69
    à¼
    -0.68
     Jem
    -0.68
     Alto
    -0.67
     Seymour
    -0.66
    olithic
    -0.65
     deceased
    -0.63
     tasting
    -0.63
     Claud
    -0.63
    POSITIVE LOGITS
    itures
    1.19
     Spending
    1.12
     expenditure
    1.09
     expenditures
    1.06
    anke
    0.96
     spending
    0.94
    iture
    0.93
     spree
    0.90
     budgets
    0.89
     Spend
    0.87
    Activations Density 0.027%

    No Known Activations