© 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. 4-RES-JB
    11. 13489
    Prev
    Next
    INDEX
    Explanations

    keywords related to a specific entity named "Watford"

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "Wat" or variations of it, likely focusing on a specific context or topic related to "Watford."

    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
     prevailing
    -0.73
    pmwiki
    -0.72
     Haas
    -0.71
     reconciliation
    -0.65
    interstitial
    -0.65
    wcs
    -0.62
     grape
    -0.62
     Sunni
    -0.61
     blunt
    -0.61
     psy
    -0.61
    POSITIVE LOGITS
    terson
    1.40
     Wat
    1.06
    chers
    1.04
    chens
    1.04
    pering
    1.02
    combe
    1.01
    cher
    1.00
    ches
    0.96
    ched
    0.95
    atche
    0.94
    Activations Density 0.007%

    No Known Activations