© 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. 8593
    Prev
    Next
    INDEX
    Explanations

    phrases related to sports events and actions

    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
    460
    +0.11
    0.5%
    652
    +0.11
    0.5%
    245
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    460
    +0.11
    0.04
    500
    +0.11
    0.04
    238
    +0.10
    0.04
    Negative Logits
    ricia
    -0.83
    apore
    -0.74
    resa
    -0.72
     actresses
    -0.71
    igmat
    -0.71
    regon
    -0.70
    anmar
    -0.70
    indust
    -0.69
     acron
    -0.69
     Canaver
    -0.69
    POSITIVE LOGITS
     victory
    1.05
     shootout
    1.01
     score
    1.00
     win
    0.99
     clinch
    0.98
     concede
    0.94
     wins
    0.93
    Result
    0.92
    Winner
    0.92
     defeat
    0.91
    Activations Density 0.298%

    No Known Activations