© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 1-RES-JB
    11. 8696
    Prev
    Next
    INDEX
    Explanations

    references to sports drafts, primarily in the NFL and NHL

    oai_token-act-pair · gpt-3.5-turbo

    references to sports drafts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tesy
    -0.76
    aunder
    -0.75
     gent
    -0.72
    virt
    -0.68
    adal
    -0.68
    none
    -0.67
    semb
    -0.66
    Sax
    -0.66
    otto
    -0.66
    asio
    -0.65
    POSITIVE LOGITS
    smanship
    1.28
    Kings
    1.23
     Draft
    1.13
    Draft
    1.01
     drafts
    0.94
    draft
    0.93
     draft
    0.89
     Combine
    0.83
     Sharks
    0.75
     Document
    0.73
    Activations Density 0.009%

    No Known Activations