© 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. 8638
    Prev
    Next
    INDEX
    Explanations

    references to arcade games

    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
    288
    +0.14
    0.6%
    549
    +0.10
    0.5%
    481
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.14
    0.02
    288
    +0.10
    0.03
    628
    +0.10
    0.02
    Negative Logits
    hillary
    -0.79
    ndra
    -0.77
    icka
    -0.77
    lying
    -0.72
     Marketable
    -0.71
    arah
    -0.70
    deals
    -0.69
    apply
    -0.68
    alez
    -0.67
    thening
    -0.67
    POSITIVE LOGITS
     Pinball
    0.99
     arcade
    0.94
     Mega
    0.93
     2600
    0.92
     emulator
    0.92
     Classics
    0.92
     Atari
    0.92
     Classic
    0.90
     Sega
    0.87
     cabinets
    0.85
    Activations Density 0.069%

    No Known Activations