© 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. 19717
    Prev
    Next
    INDEX
    Explanations

    information related to diets high in antioxidants, healthy lifestyles, and regular exercise

    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
    732
    +0.10
    0.5%
    325
    +0.10
    0.4%
    248
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    162
    +0.10
    0.03
    678
    +0.10
    0.03
    35
    +0.09
    0.03
    Negative Logits
     jurisd
    -0.87
     Torn
    -0.78
     Constable
    -0.76
     seism
    -0.75
     Franchise
    -0.75
     Inspector
    -0.74
     voic
    -0.72
     seismic
    -0.71
     Jury
    -0.71
     Letters
    -0.69
    POSITIVE LOGITS
     diet
    1.65
     diets
    1.46
     dietary
    1.31
     meals
    1.28
    ohydrate
    1.25
     calorie
    1.25
     calories
    1.25
     carbohydrates
    1.24
     foods
    1.23
     lifestyle
    1.21
    Activations Density 0.387%

    No Known Activations