© 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. 1-RES-JB
    11. 9514
    Prev
    Next
    INDEX
    Explanations

    names of locations, possibly related to a specific city or town

    oai_token-act-pair · gpt-3.5-turbo

    references to the name "Fremont" and associated terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    draw
    -0.82
     Cassidy
    -0.72
    lihood
    -0.67
    library
    -0.66
    Downloadha
    -0.64
     Twain
    -0.63
     Gerry
    -0.63
    BOOK
    -0.62
    terday
    -0.61
    ums
    -0.61
    POSITIVE LOGITS
    antle
    1.43
     Frem
    1.12
    cipled
    0.85
     corrid
    0.79
    puter
    0.79
    actory
    0.77
    bledon
    0.76
    inite
    0.76
    shire
    0.74
    galitarian
    0.74
    Activations Density 0.009%

    No Known Activations