Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 21834
    Prev
    Next
    INDEX
    Explanations

    words related to racing events or teams

    oai_token-act-pair · gpt-3.5-turbo

    references to various racing teams and competitions

    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
     typical
    -0.72
     child
    -0.72
     phase
    -0.71
     appar
    -0.69
     natural
    -0.66
     summ
    -0.66
     dough
    -0.64
     penetration
    -0.64
     stream
    -0.63
     swall
    -0.63
    POSITIVE LOGITS
     Racing
    4.14
     Motorsport
    2.04
     Motors
    1.39
     Speedway
    1.31
     Cycling
    1.30
     Sports
    1.27
     Rac
    1.27
     racing
    1.26
     racer
    1.20
     Sporting
    1.18
    Activations Density 0.013%

    No Known Activations