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. 8-RES-JB
    6. 1564
    Prev
    Next
    INDEX
    Explanations

    statistics and achievements in sports

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "with" and associated phrases indicating relationships or statistics in a sports context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    late
    -0.69
    Uk
    -0.69
    hemat
    -0.69
    ulture
    -0.68
    mite
    -0.67
    acci
    -0.66
    ultural
    -0.65
    insula
    -0.65
    strate
    -0.65
    TH
    -0.63
    POSITIVE LOGITS
    stood
    1.33
     regards
    1.13
    drawn
    1.06
     impunity
    1.05
     regard
    1.04
    draw
    0.95
    standing
    0.83
    holding
    0.82
     runners
    0.76
     respect
    0.75
    Activations Density 0.125%

    No Known Activations