© 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. 5433
    Prev
    Next
    INDEX
    Explanations

    references to historical figures and events, including Latin America during the Colonial Period and French politics during the Reign of Terror

    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
    447
    +0.12
    0.6%
    497
    +0.10
    0.4%
    494
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.12
    0.07
    755
    +0.10
    0.04
    476
    +0.10
    0.03
    Negative Logits
    outine
    -0.78
    Tickets
    -0.75
     withhold
    -0.72
     violates
    -0.72
    urden
    -0.71
    ratom
    -0.70
     violate
    -0.69
    lamm
    -0.66
     tether
    -0.66
     retweet
    -0.65
    POSITIVE LOGITS
     1960
    1.40
     1970
    1.40
     1930
    1.39
    ixties
    1.39
     1950
    1.37
     1980
    1.34
     1990
    1.33
     Era
    1.32
     1940
    1.32
     1920
    1.31
    Activations Density 0.391%

    No Known Activations