© 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. 0-RES-JB
    11. 16497
    Prev
    Next
    INDEX
    Explanations

    references to sports stadiums

    oai_token-act-pair · gpt-3.5-turbo
    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
    Edited
    -0.63
    {{
    -0.63
    Maker
    -0.63
     Mic
    -0.62
     Diss
    -0.62
    BER
    -0.62
    Lib
    -0.62
    nom
    -0.61
     Dial
    -0.60
    ect
    -0.60
    POSITIVE LOGITS
     stadium
    3.39
     stadiums
    2.76
     Stadium
    2.45
     ballpark
    1.99
     arena
    1.67
     Coliseum
    1.55
     venue
    1.50
     Wembley
    1.44
    iseum
    1.35
     cathedral
    1.33
    Activations Density 0.017%

    No Known Activations