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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 10-RES-JB
    11. 13633
    Prev
    Next
    INDEX
    Explanations

    details about sports performance, including matches played, goals scored, and teams played for

    oai_token-act-pair · gpt-3.5-turbo

    details related to a player's statistics and performance in sports

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    morning
    -0.72
     Parkinson
    -0.71
    #$
    -0.71
    erity
    -0.68
     Canaver
    -0.67
    Ħ¢
    -0.66
     Gins
    -0.66
     FDR
    -0.66
     Rockefeller
    -0.62
     Dalai
    -0.61
    POSITIVE LOGITS
     AHL
    0.92
     stint
    0.91
     USL
    0.83
     Playoffs
    0.82
     appearances
    0.82
     affiliate
    0.79
    rafted
    0.77
     registering
    0.76
    arsity
    0.76
     colleg
    0.75
    Activations Density 0.296%

    No Known Activations