© 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 EvalsExportsinterp-engineNEWAPI 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. 2334
    Prev
    Next
    INDEX
    Explanations

    comparisons between different conditions or treatments in scientific studies

    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
    651
    +0.12
    0.6%
    508
    +0.12
    0.5%
    338
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    508
    +0.12
    0.03
    651
    +0.12
    0.03
    409
    +0.10
    0.02
    Negative Logits
    atron
    -0.75
    wind
    -0.73
    peed
    -0.71
     embr
    -0.70
     Poles
    -0.66
    dust
    -0.65
    othe
    -0.65
    yrights
    -0.64
     totem
    -0.63
     Bounce
    -0.63
    POSITIVE LOGITS
     randomized
    1.10
     study
    1.10
     trials
    1.09
     Trial
    1.08
     trial
    1.03
    Study
    1.02
    clinical
    1.02
     Study
    1.02
     clinical
    1.02
    trial
    1.00
    Activations Density 0.118%

    No Known Activations