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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 9804
    Prev
    Next
    INDEX
    Explanations

    mentions of relations or interactions with friends

    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
    377
    +0.13
    0.6%
    247
    +0.12
    0.6%
    420
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    377
    +0.13
    0.04
    247
    +0.12
    0.03
    506
    +0.10
    0.04
    Negative Logits
     chloride
    -0.69
    grim
    -0.67
    imon
    -0.66
    phas
    -0.66
    tarians
    -0.65
    yss
    -0.64
    teen
    -0.62
    elsius
    -0.62
     Cout
    -0.62
     dioxide
    -0.61
    POSITIVE LOGITS
    liest
    1.34
    liness
    1.33
    lier
    1.27
    lies
    1.15
    hips
    0.98
    ships
    0.97
    ship
    0.95
    folk
    0.83
    finder
    0.79
     acquaintances
    0.77
    Activations Density 0.096%

    No Known Activations