© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 6191
    Prev
    Next
    INDEX
    Explanations

    references to archives or archived content

    oai_token-act-pair · gpt-3.5-turbo

    references to archival material or collections

    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
    nir
    -0.77
    ggle
    -0.75
    laws
    -0.71
    nen
    -0.71
    ners
    -0.70
    osuke
    -0.70
    pher
    -0.69
    yah
    -0.68
    hua
    -0.68
    acted
    -0.68
    POSITIVE LOGITS
     archive
    0.98
     archives
    0.97
     trove
    0.92
     Archive
    0.90
    conservancy
    0.87
     dumps
    0.85
    archive
    0.84
     documenting
    0.83
     collection
    0.83
    archives
    0.83
    Activations Density 0.015%

    No Known Activations