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 1536
    5. 8-RES_FS1536-JB
    6. 111
    Prev
    Next
    INDEX
    Explanations

    descriptions related to sports events, particularly boxing and UFC fights

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    346
    +0.11
    0.5%
    682
    +0.10
    0.5%
    24
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    24
    +0.11
    0.07
    652
    +0.10
    0.08
    500
    +0.10
    0.08
    Negative Logits
    Property
    -0.85
     Gutenberg
    -0.85
     Orwell
    -0.84
     UNCLASSIFIED
    -0.84
    ibrary
    -0.84
     Dickens
    -0.83
    âĹ¼
    -0.80
     Helpful
    -0.80
    taboola
    -0.79
    adobe
    -0.78
    POSITIVE LOGITS
     Championship
    1.45
     tournament
    1.40
     championship
    1.38
     rematch
    1.35
     Championships
    1.32
     Finals
    1.30
     undefeated
    1.27
     Tournament
    1.27
     finals
    1.25
     tournaments
    1.24
    Activations Density 3.733%

    No Known Activations