© 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. 6595
    Prev
    Next
    INDEX
    Explanations

    exclamatory expressions of surprise or amazement

    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
    274
    +0.10
    0.4%
    45
    +0.10
    0.4%
    377
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    572
    +0.10
    0.01
    326
    +0.10
    0.01
    377
    +0.09
    0.01
    Negative Logits
    href
    -0.71
     redress
    -0.70
     delinqu
    -0.70
    icipated
    -0.65
     externalToEVAOnly
    -0.64
     obligated
    -0.63
     kindred
    -0.62
    ":[{"
    -0.62
     subsistence
    -0.62
     fullest
    -0.61
    POSITIVE LOGITS
    zers
    1.22
     Wow
    0.98
    Wow
    0.94
     wow
    0.93
    pedia
    0.90
    orld
    0.90
    !,
    0.86
    wow
    0.86
    !:
    0.83
    ww
    0.82
    Activations Density 0.018%

    No Known Activations