© 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. 10487
    Prev
    Next
    INDEX
    Explanations

    instances of the word "Two"

    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
    748
    +0.10
    0.4%
    420
    +0.10
    0.4%
    357
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    748
    +0.10
    0.06
    357
    +0.10
    0.07
    223
    +0.09
    0.07
    Negative Logits
    ifice
    -0.77
    asta
    -0.73
    amaru
    -0.73
    needs
    -0.69
    charge
    -0.68
    ugu
    -0.68
     Caption
    -0.68
    annel
    -0.68
    ook
    -0.67
    ashtra
    -0.67
    POSITIVE LOGITS
     thirds
    1.56
     dozen
    1.18
     halves
    1.13
     weeks
    1.10
     hundred
    1.08
    fold
    1.05
    teenth
    1.02
    nd
    0.98
     separate
    0.96
     consecutive
    0.91
    Activations Density 0.185%

    No Known Activations