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 768
    5. 8-RES_FS768-JB
    6. 112
    Prev
    Next
    INDEX
    Explanations

    terms related to programming, syntax, and computational concepts

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    480
    +0.54
    4.9%
    447
    +0.38
    3.4%
    266
    +0.12
    1.1%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.54
    0.05
    266
    +0.38
    0.43
    481
    +0.12
    0.38
    Negative Logits
     burgeoning
    -0.29
     Vaugh
    -0.28
    wealth
    -0.27
    anwhile
    -0.27
    urring
    -0.27
     Newly
    -0.26
     fledgling
    -0.26
     glamorous
    -0.25
     marquee
    -0.25
     fuels
    -0.25
    POSITIVE LOGITS
     tha
    0.34
     tho
    0.32
     alot
    0.32
    tch
    0.29
     clipping
    0.29
     [+
    0.29
     kinda
    0.28
     [/
    0.28
    natureconservancy
    0.28
    pload
    0.28
    Activations Density 20.987%

    No Known Activations