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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 7811
    Prev
    Next
    INDEX
    Explanations

    references to actions in a visual scene, particularly involving people and movement

    oai_token-act-pair · gpt-3.5-turbo

    descriptions of video footage involving individuals in various scenarios

    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
    Pand
    -0.86
    ãĤ¦ãĤ¹
    -0.82
    trade
    -0.78
    profits
    -0.77
    inventoryQuantity
    -0.76
     founded
    -0.73
    manac
    -0.71
    soType
    -0.70
    ldom
    -0.70
    utics
    -0.68
    POSITIVE LOGITS
     footage
    1.10
     filming
    1.05
     clips
    1.04
     kneeling
    1.03
     filmed
    1.02
     narrator
    0.97
     camer
    0.95
     clip
    0.94
     crou
    0.91
     flashing
    0.90
    Activations Density 0.653%

    No Known Activations