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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 6200
    Prev
    Next
    INDEX
    Explanations

    names of individuals related to legal cases or disputes

    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
    348
    +0.10
    0.5%
    464
    +0.10
    0.5%
    41
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.10
    0.09
    232
    +0.10
    0.07
    41
    +0.10
    0.07
    Negative Logits
     Beacon
    -0.82
     Monroe
    -0.75
     Wilmington
    -0.74
     Tacoma
    -0.73
     Sounders
    -0.72
     Becky
    -0.70
     Abby
    -0.70
    Weiss
    -0.69
    Miami
    -0.69
     Brittany
    -0.69
    POSITIVE LOGITS
     Singh
    1.60
     Sharma
    1.52
     Kumar
    1.48
     Patel
    1.33
     Rao
    1.30
     Gupta
    1.29
     Bhar
    1.27
     Pradesh
    1.27
     Sabha
    1.26
    bh
    1.24
    Activations Density 0.413%

    No Known Activations