© 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. 10289
    Prev
    Next
    INDEX
    Explanations

    mentions of specific French food items

    oai_token-act-pair · gpt-3.5-turbo

    references to food items, particularly appetizers and the phrase "a dime a dozen."

    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
     Paige
    -0.74
     McCabe
    -0.72
     Islamabad
    -0.71
     Pesh
    -0.67
    slave
    -0.66
    keeper
    -0.66
    borg
    -0.65
    etheus
    -0.65
     Bean
    -0.64
     Origin
    -0.63
    POSITIVE LOGITS
     hors
    3.25
     dime
    2.02
     Hof
    1.18
    los
    1.09
     lou
    1.02
     dors
    0.97
     vot
    0.97
     les
    0.90
     marqu
    0.87
     amuse
    0.86
    Activations Density 0.046%

    No Known Activations