© 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. 3005
    Prev
    Next
    INDEX
    Explanations

    words related to legal or political contexts

    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
    +0.14
    0.6%
    284
    +0.10
    0.5%
    308
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.14
    0.03
    284
    +0.10
    0.02
    288
    +0.10
    0.03
    Negative Logits
    sie
    -0.99
    s
    -0.99
    son
    -0.97
    doms
    -0.88
    matically
    -0.86
    matic
    -0.83
    sin
    -0.82
    oos
    -0.80
    intosh
    -0.80
    shit
    -0.80
    POSITIVE LOGITS
     Lauder
    0.83
    lda
    0.80
    chnology
    0.71
    uthor
    0.70
    uve
    0.68
    ploy
    0.66
     Scotia
    0.66
    uner
    0.65
    ption
    0.64
    ptives
    0.64
    Activations Density 0.065%

    No Known Activations