© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 9-RES-JB
    11. 24268
    Prev
    Next
    INDEX
    Explanations

    mentions of game-related events such as winning goals, plays, or turning points

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to sports games and their outcomes

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     Lanka
    -0.76
     Burr
    -0.69
    ulhu
    -0.68
    auga
    -0.66
     nort
    -0.62
    veland
    -0.61
    fty
    -0.60
     Hai
    -0.59
     Borders
    -0.59
     cig
    -0.58
    POSITIVE LOGITS
    winning
    1.00
    saving
    0.98
    related
    0.97
    driving
    0.92
    manship
    0.91
    changing
    0.91
    driven
    0.90
    grade
    0.89
    sized
    0.89
    making
    0.89
    Activations Density 0.061%

    No Known Activations