© 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 · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 1617
    Prev
    Next
    INDEX
    Explanations

    references to sports teams, player names, and event locations

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    373
    +0.12
    0.6%
    155
    +0.09
    0.4%
    526
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.12
    0.09
    526
    +0.09
    -0.01
    547
    +0.09
    0.02
    Negative Logits
     Ibid
    -0.74
     UNCLASSIFIED
    -0.74
     ®
    -0.73
    )",
    -0.71
     constitu
    -0.68
     Citiz
    -0.67
    SOURCE
    -0.65
    ebin
    -0.63
    swers
    -0.61
     NGOs
    -0.60
    POSITIVE LOGITS
    ESPN
    1.08
     Stadium
    1.03
     jerseys
    0.95
     ESPN
    0.95
     matchup
    0.94
     Coliseum
    0.92
     Sports
    0.91
    sports
    0.90
     Athletic
    0.90
    Football
    0.89
    Activations Density 0.717%

    No Known Activations