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

    Neuronpedia

    APICircuit 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. 0-RES-JB
    6. 10638
    Prev
    Next
    INDEX
    Explanations

    instances of the word "barrels"

    oai_token-act-pair · gpt-3.5-turbo

    references to quantities of barrels and containers related to oil production

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     glimpse
    -0.86
     piece
    -0.85
     job
    -0.81
     strand
    -0.80
     disciple
    -0.78
     contender
    -0.77
    ringe
    -0.77
     member
    -0.77
    obook
    -0.77
     fragment
    -0.76
    POSITIVE LOGITS
     crates
    1.89
     barrels
    1.84
     desks
    1.81
     chests
    1.80
     containers
    1.79
     tanks
    1.75
     carts
    1.73
     boxes
    1.73
     detectors
    1.73
     jars
    1.72
    Activations Density 0.124%

    No Known Activations