© 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. 3738
    Prev
    Next
    INDEX
    Explanations

    mentions of physical barriers or architecture, especially in the context of construction, restoration, or historical significance

    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
    325
    +0.10
    0.5%
    375
    +0.10
    0.5%
    288
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    325
    +0.10
    0.03
    99
    +0.10
    0.02
    737
    +0.10
    0.02
    Negative Logits
    Gene
    -0.86
    Pacific
    -0.72
    phrine
    -0.71
    lihood
    -0.70
     delegated
    -0.68
     Eug
    -0.65
    MENTS
    -0.65
    orean
    -0.64
    milo
    -0.64
    Wild
    -0.63
    POSITIVE LOGITS
    papers
    1.26
    abies
    1.14
    aby
    1.06
     walls
    0.96
     wall
    0.94
    paper
    0.94
    clock
    0.92
     thickness
    0.89
    buster
    0.89
    top
    0.87
    Activations Density 0.021%

    No Known Activations