© 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. 24042
    Prev
    Next
    INDEX
    Explanations

    terms related to demands, promises, and orders

    oai_token-act-pair · gpt-4o-miniTriggered 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
    224
    +0.14
    0.6%
    739
    +0.10
    0.4%
    16
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    540
    +0.14
    0.05
    224
    +0.10
    0.04
    247
    +0.10
    0.03
    Negative Logits
    allery
    -0.85
    ocks
    -0.79
    vae
    -0.77
    verage
    -0.74
    aples
    -0.73
     Coverage
    -0.71
    Studies
    -0.70
    iencies
    -0.70
    aces
    -0.69
    culus
    -0.67
    POSITIVE LOGITS
     letter
    0.93
     fulfilled
    0.89
     forb
    0.86
     granted
    0.86
     proclamation
    0.85
     granting
    0.84
    naires
    0.83
     issued
    0.82
     decree
    0.82
     pardon
    0.82
    Activations Density 0.130%

    No Known Activations