© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 8164
    Prev
    Next
    INDEX
    Explanations

    technical terms related to software development frameworks and systems

    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
    288
    +0.17
    0.8%
    447
    +0.14
    0.6%
    679
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    435
    +0.17
    0.01
    288
    +0.14
    0.01
    377
    +0.13
    0.01
    Negative Logits
    ucks
    -0.89
    wagen
    -0.80
    hur
    -0.77
    frey
    -0.75
    yss
    -0.71
    yd
    -0.69
    nen
    -0.69
    abouts
    -0.67
    wage
    -0.67
    uck
    -0.67
    POSITIVE LOGITS
     interfaces
    1.04
     interface
    0.96
    eering
    0.82
     layout
    0.82
     mund
    0.82
     elements
    0.81
     layouts
    0.79
     designers
    0.78
     design
    0.78
     Builder
    0.78
    Activations Density 0.013%

    No Known Activations