© 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 12288
    9. 8-RES_FS12288-JB
    11. 10684
    Prev
    Next
    INDEX
    Explanations

    names of specific individuals or brands

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    634
    +0.11
    0.5%
    526
    +0.11
    0.5%
    641
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    634
    +0.11
    0.02
    326
    +0.11
    0.02
    0
    +0.10
    0.02
    Negative Logits
    owship
    -0.74
    stood
    -0.73
     narrator
    -0.70
    UAL
    -0.69
     EDITION
    -0.69
    DERR
    -0.67
    ATIONAL
    -0.65
    lain
    -0.64
    worthiness
    -0.64
     embodiments
    -0.62
    POSITIVE LOGITS
    pper
    1.27
    ppy
    1.20
    pping
    1.20
    etry
    1.17
    achers
    1.13
    orer
    1.11
    ppa
    1.11
    ppers
    1.08
    aching
    1.07
    ppel
    1.06
    Activations Density 0.081%

    No Known Activations