© 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. 19360
    Prev
    Next
    INDEX
    Explanations

    sports-related terms, specifically related to soccer

    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
    232
    +0.12
    0.5%
    30
    +0.10
    0.4%
    240
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.12
    0.02
    659
    +0.10
    0.02
    30
    +0.09
    0.01
    Negative Logits
    ãĥ¥
    -0.71
    vez
    -0.70
    cean
    -0.68
    Trade
    -0.64
    cephal
    -0.64
    chio
    -0.64
    ITAL
    -0.63
     kidney
    -0.62
     partial
    -0.61
     saline
    -0.61
    POSITIVE LOGITS
    iday
    1.25
    idays
    1.10
    enberg
    1.00
    ican
    0.98
    ington
    0.97
    aday
    0.92
    igan
    0.92
    inger
    0.92
    oman
    0.92
    esley
    0.89
    Activations Density 0.033%

    No Known Activations