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. 21244
    Prev
    Next
    INDEX
    Explanations

    references to sports channels or shows

    oai_token-act-pair · gpt-3.5-turbo

    mentions of sports-related entities and names

    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
     encamp
    -0.69
     sequence
    -0.69
     dwell
    -0.67
     plot
    -0.66
     lid
    -0.65
     supposed
    -0.64
     disarm
    -0.62
     code
    -0.62
     heroine
    -0.62
     basil
    -0.61
    POSITIVE LOGITS
    Sports
    3.45
    Sport
    1.77
    Football
    1.72
    sports
    1.67
    Gi
    1.30
    Professional
    1.25
    Soc
    1.14
    Stars
    1.10
    Gaming
    1.10
    Coach
    1.07
    Activations Density 0.019%

    No Known Activations