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 3072
    5. 8-RES_FS3072-JB
    6. 490
    Prev
    Next
    INDEX
    Explanations

    information related to sports, including player injuries, team decisions, and player statistics

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    625
    +0.10
    0.5%
    761
    +0.09
    0.4%
    538
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.10
    0.14
    232
    +0.09
    0.10
    538
    +0.09
    0.07
    Negative Logits
     Feminist
    -0.86
    netflix
    -0.84
    DragonMagazine
    -0.83
    uscript
    -0.79
    enegger
    -0.77
     Rowling
    -0.74
     Cosponsors
    -0.73
    indust
    -0.71
    ombat
    -0.71
    Operation
    -0.70
    POSITIVE LOGITS
     teammate
    1.15
     rookie
    1.08
     Rookie
    1.03
     linebacker
    1.02
     QB
    1.01
     quarterback
    1.00
     Manziel
    0.99
     undrafted
    0.99
     injury
    0.99
     fullback
    0.98
    Activations Density 1.044%

    No Known Activations