© 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. 15449
    Prev
    Next
    INDEX
    Explanations

    technical terms related to computing and technology

    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
    311
    +0.11
    0.5%
    640
    +0.11
    0.5%
    44
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.11
    0.03
    23
    +0.11
    0.02
    311
    +0.09
    0.01
    Negative Logits
     certs
    -0.73
     wonders
    -0.71
     pores
    -0.71
    boards
    -0.65
    FontSize
    -0.61
     rattled
    -0.61
     lett
    -0.61
    daq
    -0.60
     nowhere
    -0.59
    lves
    -0.59
    POSITIVE LOGITS
     Firstly
    1.23
     namely
    0.95
    Firstly
    0.89
     Either
    0.80
     First
    0.80
    hemat
    0.74
     Ones
    0.72
    first
    0.71
     viz
    0.68
     secondly
    0.67
    Activations Density 0.112%

    No Known Activations