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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 19326
    Prev
    Next
    INDEX
    Explanations

    years mentioned in the context of future plans or forecasts

    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
    260
    +0.12
    0.6%
    314
    +0.11
    0.5%
    652
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    223
    +0.12
    0.04
    260
    +0.11
    0.04
    693
    +0.11
    0.03
    Negative Logits
     envy
    -0.73
    TextColor
    -0.66
     mockery
    -0.60
     leveled
    -0.60
     scissors
    -0.60
     tender
    -0.59
     Apostle
    -0.58
     Selection
    -0.58
    ogyn
    -0.57
     Telegram
    -0.56
    POSITIVE LOGITS
    mins
    0.94
     minutes
    0.92
    %.
    0.92
     mins
    0.91
    km
    0.91
     kilometres
    0.90
     hours
    0.86
     milliseconds
    0.85
     months
    0.84
    th
    0.84
    Activations Density 0.080%

    No Known Activations