© 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. 20110
    Prev
    Next
    INDEX
    Explanations

    mentions of sports teams (Watford) and related activities

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Wat" and its variations, likely related to a specific theme or subject pertaining to Watford

    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
     Haas
    -0.74
    interstitial
    -0.71
     prevailing
    -0.68
    pmwiki
    -0.66
     affirmative
    -0.62
     blunt
    -0.61
     appropriation
    -0.61
     lobbying
    -0.60
    EStream
    -0.59
    sharing
    -0.59
    POSITIVE LOGITS
    terson
    1.33
     Wat
    1.27
    combe
    0.99
     wat
    0.93
    chers
    0.93
    auga
    0.90
    chens
    0.90
    chet
    0.89
    ako
    0.86
    cher
    0.86
    Activations Density 0.004%

    No Known Activations