© 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. 6584
    Prev
    Next
    INDEX
    Explanations

    photo credits

    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
    373
    +0.15
    0.7%
    441
    +0.11
    0.5%
    109
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    323
    +0.15
    0.01
    513
    +0.11
    0.01
    424
    +0.09
    0.01
    Negative Logits
    velt
    -0.77
    iqueness
    -0.73
     tyres
    -0.66
    ãĤ¦ãĤ¹
    -0.65
     tires
    -0.63
    otype
    -0.62
     intest
    -0.62
    ullivan
    -0.62
    ailability
    -0.61
    ãĤ¡
    -0.61
    POSITIVE LOGITS
     Shutterstock
    0.89
     SPL
    0.73
    EDIT
    0.72
     EDIT
    0.71
     Contribut
    0.69
    ="/
    0.69
     ©
    0.68
     Provided
    0.68
     Attribution
    0.67
     Uriel
    0.67
    Activations Density 0.021%

    No Known Activations