© 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 EvalsExportsAPI 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. 20129
    Prev
    Next
    INDEX
    Explanations

    sentences related to political commentary and societal critique

    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
    756
    +0.21
    1.1%
    87
    +0.18
    0.9%
    742
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    756
    +0.21
    -0.00
    390
    +0.18
    0.03
    87
    +0.09
    -0.04
    Negative Logits
     Emails
    -0.73
     tracker
    -0.67
    peed
    -0.66
     Clicker
    -0.66
     Reload
    -0.66
    cheat
    -0.66
     Radar
    -0.66
     PLEASE
    -0.62
     @
    -0.62
     Collider
    -0.62
    POSITIVE LOGITS
     colonialism
    1.13
     centuries
    1.03
     millennia
    1.02
     globalization
    1.02
     civilizations
    1.00
     generations
    0.98
     emanc
    0.97
     industrialized
    0.96
     postwar
    0.96
     democracies
    0.96
    Activations Density 1.221%

    No Known Activations