© 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. 1282
    Prev
    Next
    INDEX
    Explanations

    phrases related to dietary restrictions and food options, specifically focusing on gluten and dairy-free options

    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
    481
    +0.17
    0.8%
    421
    +0.12
    0.6%
    289
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    289
    +0.17
    0.01
    285
    +0.12
    0.01
    57
    +0.12
    0.01
    Negative Logits
    CLASSIFIED
    -0.80
    hare
    -0.69
    ILA
    -0.68
    pload
    -0.65
    swick
    -0.65
    HCR
    -0.64
    uscript
    -0.64
     mathemat
    -0.64
    jun
    -0.62
     Sent
    -0.62
    POSITIVE LOGITS
     gluten
    1.02
     intolerance
    0.97
     intoler
    0.95
     allergy
    0.94
     sensitivity
    0.87
     allergies
    0.86
     digestion
    0.83
    iness
    0.83
     tolerant
    0.82
     tolerance
    0.82
    Activations Density 0.017%

    No Known Activations