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. 8-RES-JB
    6. 23165
    Prev
    Next
    INDEX
    Explanations

    terms related to government policies, regulations, and programs, particularly related to investments, legal matters, and security enforcement

    oai_token-act-pair · gpt-3.5-turbo

    terms related to governmental and financial programs or policies

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    inventoryQuantity
    -0.94
     spoilers
    -0.73
    stellar
    -0.70
    magic
    -0.68
    wikipedia
    -0.68
     quartz
    -0.67
    iasm
    -0.66
    yond
    -0.66
    xus
    -0.66
    strings
    -0.65
    POSITIVE LOGITS
     Association
    1.34
     Facility
    1.27
     Initiative
    1.27
     Program
    1.26
     Act
    1.25
     Fund
    1.25
     Center
    1.22
     Complex
    1.18
     Improvement
    1.16
     Conference
    1.15
    Activations Density 0.229%

    No Known Activations