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

    Neuronpedia

    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 49152
    9. 8-RES_FS49152-JB
    11. 26786
    Prev
    Next
    INDEX
    Explanations

    references to political campaigns and advertising strategies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    723
    +0.14
    0.6%
    217
    +0.13
    0.6%
    174
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    723
    +0.14
    0.03
    217
    +0.13
    0.02
    679
    +0.12
    0.04
    Negative Logits
    borg
    -0.87
    elsius
    -0.77
     decom
    -0.77
     physicists
    -0.76
     citiz
    -0.74
    ucl
    -0.73
     Centauri
    -0.68
     Biological
    -0.68
    poral
    -0.68
     ruled
    -0.67
    POSITIVE LOGITS
    campaign
    1.28
     fundraising
    1.12
     touting
    1.10
     campaigns
    1.10
     endorsements
    1.05
     billboards
    1.05
     campaign
    1.05
     ads
    1.05
     messaging
    1.03
    Campaign
    1.03
    Activations Density 0.221%

    No Known Activations