© 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. 17505
    Prev
    Next
    INDEX
    Explanations

    mentions of political and government-related terms and events

    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
    291
    +0.09
    0.4%
    710
    +0.09
    0.4%
    766
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    291
    +0.09
    0.05
    1
    +0.09
    0.04
    710
    +0.09
    0.04
    Negative Logits
    iculture
    -0.90
    otor
    -0.86
     ç¥ŀ
    -0.82
    oba
    -0.77
    Interstitial
    -0.77
    icultural
    -0.76
    oka
    -0.74
    WIN
    -0.69
    entary
    -0.68
     stabil
    -0.68
    POSITIVE LOGITS
     leaked
    1.50
     revelations
    1.47
     whistleblower
    1.41
     disclosures
    1.34
     whistleblowers
    1.27
     allegations
    1.26
     WikiLeaks
    1.24
     incrim
    1.24
     whistlebl
    1.23
     leaks
    1.22
    Activations Density 0.708%

    No Known Activations