© 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. 3234
    Prev
    Next
    INDEX
    Explanations

    information relating to public demonstrations, protests, political activities, and international tensions

    oai_token-act-pair · gpt-3.5-turbo
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.16
    0.7%
    481
    +0.12
    0.5%
    608
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.16
    0.06
    608
    +0.12
    0.06
    616
    +0.10
    0.04
    Negative Logits
     Norn
    -0.81
    orthy
    -0.76
     buddies
    -0.75
    Favorite
    -0.69
     mates
    -0.69
    minus
    -0.68
    icator
    -0.68
    wright
    -0.67
     wisely
    -0.66
     Gabe
    -0.65
    POSITIVE LOGITS
     uproar
    1.52
     controversy
    1.51
     outcry
    1.48
     backlash
    1.41
     upheaval
    1.36
     controversies
    1.29
     turmoil
    1.29
     protests
    1.28
     clashes
    1.27
     crackdown
    1.26
    Activations Density 0.364%

    No Known Activations