© 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. 1-RES-JB
    11. 14162
    Prev
    Next
    INDEX
    Explanations

    phrases related to combat sports such as UFC and specific weight classes or individual fighters

    oai_token-act-pair · gpt-3.5-turbo

    terms related to combat sports and athletic achievements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    mberg
    -0.80
    etheless
    -0.71
    bear
    -0.66
    arton
    -0.64
    haar
    -0.63
     Curiosity
    -0.63
     Petition
    -0.63
    uld
    -0.62
    vironment
    -0.62
     ESA
    -0.62
    POSITIVE LOGITS
     Boxing
    1.01
     boxer
    1.01
     rematch
    0.96
     heavyweight
    0.93
     fights
    0.91
     boxing
    0.91
     wrestlers
    0.87
    Fighting
    0.87
     promoter
    0.86
     Mayweather
    0.86
    Activations Density 0.236%

    No Known Activations