Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 11139
    Prev
    Next
    INDEX
    Explanations

    descriptions and discussions related to careers and developments of athletes, particularly in the context of sports teams

    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
    373
    +0.10
    0.5%
    668
    +0.10
    0.5%
    429
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.10
    0.06
    603
    +0.10
    0.04
    761
    +0.09
    0.05
    Negative Logits
    SPONSORED
    -0.71
     Suppose
    -0.70
    agate
    -0.69
     Sec
    -0.66
    ··
    -0.65
    ÅĤ
    -0.63
     âĨij
    -0.62
     Thou
    -0.61
     Delete
    -0.61
     *.
    -0.60
    POSITIVE LOGITS
     teammates
    1.20
     teammate
    1.05
     linem
    1.03
     rehab
    1.01
     himself
    0.96
     offseason
    0.95
     fullback
    0.94
     preseason
    0.93
     rookie
    0.93
    ESPN
    0.91
    Activations Density 0.517%

    No Known Activations