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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 4-RES-JB
    6. 21088
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal or political terms

    oai_token-act-pair · gpt-3.5-turbo

    references to the Indian Reorganization Act (IRA) and related terms

    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
    Oracle
    -0.81
    loop
    -0.81
    deck
    -0.79
    opolis
    -0.76
    space
    -0.74
    package
    -0.72
    :{
    -0.67
    ifiers
    -0.67
    items
    -0.67
    maps
    -0.67
    POSITIVE LOGITS
     IRA
    0.89
     Klux
    0.81
     graz
    0.73
    agh
    0.70
    ISH
    0.70
     trustee
    0.69
    leans
    0.69
    zzi
    0.68
    eed
    0.68
     withdrawals
    0.68
    Activations Density 0.005%

    No Known Activations