© 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. 1349
    Prev
    Next
    INDEX
    Explanations

    statements related to policies and regulations, particularly in academic or research settings

    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
    46
    +0.10
    0.5%
    730
    +0.09
    0.4%
    23
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    730
    +0.10
    0.03
    46
    +0.09
    0.02
    65
    +0.09
    0.02
    Negative Logits
    elines
    -0.77
    mast
    -0.75
    itiveness
    -0.69
     quartz
    -0.65
     lash
    -0.65
    wik
    -0.65
     headlights
    -0.65
    tons
    -0.64
    iasm
    -0.63
    flush
    -0.63
    POSITIVE LOGITS
     Berkeley
    0.91
     College
    0.90
    College
    0.84
     Faculty
    0.84
    urdue
    0.81
     Colleges
    0.79
     Riverside
    0.79
     Birmingham
    0.78
     Canberra
    0.78
    inguished
    0.76
    Activations Density 0.082%

    No Known Activations