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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 23459
    Prev
    Next
    INDEX
    Explanations

    terms related to sports measurements, particularly in football

    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
    448
    +0.12
    0.5%
    462
    +0.12
    0.5%
    365
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    462
    +0.12
    0.03
    162
    +0.12
    0.02
    448
    +0.09
    0.02
    Negative Logits
    chool
    -0.82
    VPN
    -0.77
     Reborn
    -0.74
    aido
    -0.72
    rology
    -0.71
    cius
    -0.70
    APS
    -0.69
    ĸļ
    -0.68
     consulted
    -0.67
    asia
    -0.66
    POSITIVE LOGITS
     homer
    0.80
     halftime
    0.77
     conversion
    0.75
     popup
    0.75
     itch
    0.73
     plea
    0.71
     alley
    0.70
     comeback
    0.70
     plateau
    0.70
     interval
    0.70
    Activations Density 0.041%

    No Known Activations