© 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. 8778
    Prev
    Next
    INDEX
    Explanations

    sports-related information, particularly focusing on team statistics and game recaps

    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
    507
    +0.12
    0.6%
    314
    +0.11
    0.5%
    615
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    507
    +0.12
    0.05
    689
    +0.11
    0.05
    220
    +0.10
    0.04
    Negative Logits
     crowds
    -0.61
     exchange
    -0.60
     exchanges
    -0.59
    ulture
    -0.58
     disturbance
    -0.58
     management
    -0.57
    Else
    -0.57
     hierarchy
    -0.56
     usefulness
    -0.56
    prises
    -0.55
    POSITIVE LOGITS
    years
    1.02
    inches
    0.98
    ½
    0.94
     inches
    0.93
    cm
    0.91
     dimensional
    0.91
    ft
    0.88
    months
    0.88
    een
    0.86
    teenth
    0.85
    Activations Density 0.144%

    No Known Activations