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. 1013
    Prev
    Next
    INDEX
    Explanations

    references to the Cleveland Browns football team

    oai_token-act-pair · gpt-3.5-turbo

    references to the Cleveland Browns, particularly in relation to their performance and notable players

    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
     sur
    -0.73
     Ul
    -0.70
    uthor
    -0.69
     SPI
    -0.69
    Link
    -0.66
     Esp
    -0.65
     paras
    -0.62
     Sven
    -0.62
    isi
    -0.62
     warp
    -0.62
    POSITIVE LOGITS
     Browns
    3.80
     Bengals
    1.90
     Cavs
    1.82
     Cavaliers
    1.76
     Cleveland
    1.75
     Steelers
    1.72
    Brown
    1.70
     Colts
    1.65
     Manziel
    1.58
     Bears
    1.58
    Activations Density 0.012%

    No Known Activations