© 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. 20897
    Prev
    Next
    INDEX
    Explanations

    references to political figures, especially those holding high-ranking positions such as Vice President or President

    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
    328
    +0.10
    0.5%
    24
    +0.10
    0.4%
    558
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    371
    +0.10
    0.03
    204
    +0.10
    0.02
    224
    +0.10
    0.02
    Negative Logits
     goats
    -0.68
     cows
    -0.66
    MpServer
    -0.61
    é¾
    -0.60
     mos
    -0.59
     plate
    -0.58
     exting
    -0.57
     reservoirs
    -0.57
     magnitude
    -0.56
    Score
    -0.56
    POSITIVE LOGITS
     Biden
    1.04
    OTUS
    0.94
     Cheney
    0.77
    ial
    0.73
     Lyndon
    0.73
    president
    0.73
     Gore
    0.72
     Pence
    0.71
    communications
    0.70
    owitz
    0.70
    Activations Density 0.035%

    No Known Activations