© 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. 3337
    Prev
    Next
    INDEX
    Explanations

    phrases related to social or political issues and calls for action

    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
    608
    +0.14
    0.6%
    104
    +0.12
    0.6%
    665
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.14
    0.07
    665
    +0.12
    0.07
    718
    +0.12
    0.06
    Negative Logits
    breaks
    -0.72
    furt
    -0.70
    rings
    -0.69
    fortunately
    -0.69
    mails
    -0.66
    rament
    -0.65
     ain
    -0.64
    lets
    -0.64
    quickShipAvailable
    -0.64
    points
    -0.63
    POSITIVE LOGITS
     abolish
    1.15
     revise
    1.14
     legalize
    1.10
     reduce
    1.10
     improve
    1.09
     extend
    1.08
     eliminate
    1.07
     engage
    1.07
     provide
    1.06
     explore
    1.06
    Activations Density 0.209%

    No Known Activations