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. 11-RES-JB
    6. 3071
    Prev
    Next
    INDEX
    Explanations

    sports-related information, including team names, rankings, matches, and results

    oai_token-act-pair · gpt-3.5-turbo

    references to teams or groups in a competitive context, particularly regarding sports

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    soType
    -0.96
     Cosponsors
    -0.68
     characterization
    -0.67
     whistlebl
    -0.63
    minist
    -0.63
    indust
    -0.63
    Adds
    -0.63
     Spiegel
    -0.62
    speaking
    -0.62
     messenger
    -0.62
    POSITIVE LOGITS
     trailed
    1.32
     lost
    1.19
     defeated
    1.11
     clin
    1.11
     beat
    1.08
     overcame
    1.06
     lose
    1.05
     finished
    1.04
     struggled
    1.04
     clinch
    1.02
    Activations Density 0.181%

    No Known Activations