© 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. 24205
    Prev
    Next
    INDEX
    Explanations

    queries seeking opinions or assessments regarding events or issues

    oai_token-act-pair · gpt-4o-miniTriggered 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
    87
    +0.45
    2.5%
    447
    +0.30
    1.7%
    64
    +0.08
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.45
    -0.01
    8
    +0.30
    0.06
    608
    +0.08
    0.08
    Negative Logits
    )</
    -0.64
    KNOWN
    -0.62
    sed
    -0.62
     afterwards
    -0.56
     guiActiveUnfocused
    -0.55
    Ingredients
    -0.55
    Contents
    -0.54
     beforehand
    -0.53
    ().
    -0.52
    FG
    -0.51
    POSITIVE LOGITS
     Canaver
    0.76
     DeVos
    0.73
     policymakers
    0.67
     progressives
    0.67
     Chomsky
    0.66
     politic
    0.65
     foregoing
    0.63
    hillary
    0.61
     Goodell
    0.61
     advocates
    0.61
    Activations Density 1.473%

    No Known Activations