© 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. 17938
    Prev
    Next
    INDEX
    Explanations

    architectural terms or references

    oai_token-act-pair · gpt-3.5-turbo

    references to an architectural structure named "Arch."

    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
     sugar
    -0.71
     conditioning
    -0.67
     wellness
    -0.66
     competitiveness
    -0.66
     fueling
    -0.64
    ween
    -0.63
     simulator
    -0.63
    erella
    -0.61
    igun
    -0.60
    ooo
    -0.60
    POSITIVE LOGITS
     Arch
    3.92
     arch
    2.37
    Arch
    2.26
    arch
    1.62
     Archbishop
    1.46
     Archives
    1.37
     Architect
    1.32
    ARCH
    1.22
     Archive
    1.20
     Archangel
    1.19
    Activations Density 0.011%

    No Known Activations