© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 45
    Prev
    Next
    INDEX
    Explanations

    descriptive words related to political and social issues

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    393
    +0.23
    1.1%
    362
    +0.22
    1.1%
    385
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.23
    0.21
    393
    +0.22
    0.13
    636
    +0.11
    0.10
    Negative Logits
     suppose
    -0.78
     namely
    -0.74
    thood
    -0.74
    SPONSORED
    -0.69
    pers
    -0.67
    manship
    -0.67
    craft
    -0.66
    bourg
    -0.65
    Iterator
    -0.65
    perse
    -0.64
    POSITIVE LOGITS
     same
    1.46
     forefront
    1.27
     slightest
    1.17
     brink
    1.15
     proverbial
    1.11
     outset
    1.11
     outskirts
    1.10
     periphery
    1.10
     sidelines
    1.09
     opposite
    1.07
    Activations Density 1.351%

    No Known Activations