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

    Neuronpedia

    Jacobian LensNEW
    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. 45
    Prev
    Next
    INDEX
    Explanations

    descriptions related to fashion and clothing choices

    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
    480
    +0.19
    0.9%
    459
    +0.10
    0.5%
    526
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    526
    +0.19
    -0.00
    351
    +0.10
    0.04
    241
    +0.10
    0.04
    Negative Logits
     Reconstruction
    -0.85
     proposed
    -0.69
     methodological
    -0.69
     Shad
    -0.69
     embry
    -0.67
     merged
    -0.67
     affidav
    -0.65
     reconstruction
    -0.65
     quotations
    -0.64
     Journalism
    -0.64
    POSITIVE LOGITS
     enjoyment
    0.96
     crave
    0.90
     hassle
    0.85
     enjoy
    0.81
     outdoors
    0.78
    gettable
    0.78
     indulge
    0.76
     stylish
    0.76
    Enjoy
    0.75
    uggage
    0.74
    Activations Density 0.861%

    No Known Activations