© 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 EvalsExportsinterp-engineNEWAPI 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. 16456
    Prev
    Next
    INDEX
    Explanations

    countries, political entities, and diplomatic actions related to global geopolitics

    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
    608
    +0.12
    0.6%
    481
    +0.11
    0.5%
    102
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.12
    0.05
    362
    +0.11
    0.04
    150
    +0.10
    0.03
    Negative Logits
    sov
    -0.73
    ics
    -0.69
    nin
    -0.66
    ksh
    -0.66
    burst
    -0.65
    ocument
    -0.65
    ser
    -0.64
    mite
    -0.63
    nosis
    -0.63
    fitted
    -0.62
    POSITIVE LOGITS
     regard
    1.48
     regards
    1.48
     respect
    1.11
    stood
    1.03
     impunity
    1.02
    draw
    0.89
     creditors
    0.87
     stakeholders
    0.78
     superiors
    0.77
     neighbours
    0.76
    Activations Density 0.151%

    No Known Activations