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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 22347
    Prev
    Next
    INDEX
    Explanations

    references to referees and officiating in sports

    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
    104
    +0.10
    0.5%
    627
    +0.10
    0.5%
    538
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    538
    +0.10
    0.01
    288
    +0.10
    0.02
    104
    +0.09
    0.01
    Negative Logits
     Flavoring
    -0.81
    DonaldTrump
    -0.81
    igslist
    -0.78
    izoph
    -0.78
    FINEST
    -0.75
    stem
    -0.73
    VILLE
    -0.69
     Growth
    -0.67
     Prosper
    -0.67
    growth
    -0.66
    POSITIVE LOGITS
     referees
    1.18
     referee
    1.15
    eree
    1.09
     offic
    0.91
     announ
    0.84
     stopp
    0.82
    keepers
    0.80
     rigged
    0.79
    keeper
    0.78
     announcer
    0.77
    Activations Density 0.062%

    No Known Activations