© 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. 12237
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "Curry" in a context related to a person or a dish

    oai_token-act-pair · gpt-3.5-turbo

    references to the name "Curry" in various contexts

    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
     bed
    -0.67
     matt
    -0.67
     mattress
    -0.66
     hex
    -0.66
     LP
    -0.65
     magnet
    -0.64
    device
    -0.64
    wealth
    -0.63
    ,—
    -0.62
     pen
    -0.61
    POSITIVE LOGITS
     Curry
    4.04
     curry
    1.96
     Durant
    1.38
    urry
    1.24
     Rice
    1.23
     Warriors
    1.22
     Cavs
    1.20
     Harden
    1.20
     Kerr
    1.19
     Crab
    1.14
    Activations Density 0.023%

    No Known Activations