© 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. 24382
    Prev
    Next
    INDEX
    Explanations

    references to scientific research and publications

    oai_token-act-pair · gpt-4o-miniTriggered 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
    640
    +0.25
    1.2%
    373
    +0.13
    0.6%
    615
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.25
    0.09
    94
    +0.13
    0.07
    539
    +0.10
    0.05
    Negative Logits
    atism
    -0.94
    abouts
    -0.82
    lections
    -0.75
    offic
    -0.70
     wardrobe
    -0.69
     vigilante
    -0.66
    minist
    -0.64
     whine
    -0.64
     enforce
    -0.64
     bung
    -0.63
    POSITIVE LOGITS
    Researchers
    1.27
    Explore
    1.17
    Scientists
    1.12
    Abstract
    1.01
     Researchers
    0.99
     Molecular
    0.97
    Study
    0.96
    RESULTS
    0.93
     Scientists
    0.93
    Research
    0.88
    Activations Density 0.219%

    No Known Activations