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. 4-RES-JB
    6. 8290
    Prev
    Next
    INDEX
    Explanations

    phrases related to racing events

    oai_token-act-pair · gpt-3.5-turbo

    references to competitive racing events, specifically those labeled as "Grand Prix."

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

    No Comments

    Negative Logits
    itable
    -0.78
    ded
    -0.76
    itably
    -0.72
    oks
    -0.71
    ayers
    -0.70
    taboola
    -0.66
    uman
    -0.66
    ows
    -0.66
    rew
    -0.64
    gren
    -0.64
    POSITIVE LOGITS
     Prix
    1.44
     Racer
    0.90
     Rac
    0.80
     Weekend
    0.78
    naire
    0.78
     racing
    0.76
     Showdown
    0.76
     Tournament
    0.76
     racer
    0.75
     Tour
    0.73
    Activations Density 0.008%

    No Known Activations