© 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. 6433
    Prev
    Next
    INDEX
    Explanations

    phrases related to contracts, professional sports, and personal decisions

    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
    645
    +0.13
    0.6%
    688
    +0.11
    0.5%
    468
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    582
    +0.13
    0.04
    240
    +0.11
    0.04
    715
    +0.10
    0.04
    Negative Logits
    PDF
    -0.74
     neatly
    -0.66
     rife
    -0.66
     ®
    -0.66
     Amid
    -0.64
     awaiting
    -0.63
    ashington
    -0.62
     penned
    -0.62
    Synopsis
    -0.61
     Contains
    -0.61
    POSITIVE LOGITS
    entimes
    1.22
     somebody
    1.03
     everybody
    0.96
     [
    0.88
     anybody
    0.86
     guys
    0.86
     hindsight
    0.86
    initely
    0.85
     guy
    0.83
    laughs
    0.82
    Activations Density 0.353%

    No Known Activations