© 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. 0-RES-JB
    11. 18070
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "crude" with varying levels of activation

    oai_token-act-pair · gpt-3.5-turbo

    references to crude oil

    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
     Immunity
    -0.86
    igi
    -0.78
    join
    -0.77
    drive
    -0.74
    itness
    -0.70
    fight
    -0.70
    Guard
    -0.68
    mentation
    -0.64
    iversary
    -0.64
    Join
    -0.63
    POSITIVE LOGITS
     crude
    3.67
     coarse
    1.32
     refined
    1.26
     gasoline
    1.25
     petroleum
    1.24
     oil
    1.23
    Oil
    1.19
     rudimentary
    1.15
     vulgar
    1.10
     refinery
    1.09
    Activations Density 0.024%

    No Known Activations