© 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 · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 14710
    Prev
    Next
    INDEX
    Explanations

    names of countries, regions, and events

    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
    481
    +0.14
    0.6%
    367
    +0.09
    0.4%
    692
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    65
    +0.14
    0.04
    406
    +0.09
    0.04
    643
    +0.09
    0.03
    Negative Logits
     McConnell
    -0.91
     Louisville
    -0.71
    ulty
    -0.71
    Reilly
    -0.70
    Denver
    -0.69
     Ellison
    -0.69
    DERR
    -0.67
     Shelby
    -0.66
    Connor
    -0.65
     Vick
    -0.65
    POSITIVE LOGITS
    ESE
    1.30
    ese
    1.28
    istan
    1.23
     Airways
    1.22
    alam
    1.01
     Airlines
    0.99
     nationals
    0.95
    ian
    0.95
     annexed
    0.94
    utan
    0.94
    Activations Density 0.227%

    No Known Activations