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. 21856
    Prev
    Next
    INDEX
    Explanations

    descriptions of projects or activities related to personal interests and experiences

    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
    640
    +0.16
    0.7%
    679
    +0.12
    0.6%
    280
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.16
    0.12
    679
    +0.12
    0.10
    145
    +0.12
    0.09
    Negative Logits
     aven
    -0.73
     cameo
    -0.73
     disemb
    -0.72
     utter
    -0.71
    ozyg
    -0.70
     repe
    -0.69
     mysteriously
    -0.69
    plet
    -0.69
     redeemed
    -0.68
     hears
    -0.68
    POSITIVE LOGITS
     Ideally
    1.69
     Otherwise
    1.56
     Unfortunately
    1.42
     Fortunately
    1.42
     Luckily
    1.41
     Thankfully
    1.36
     Specifically
    1.27
     Especially
    1.25
     Doing
    1.21
     Instead
    1.21
    Activations Density 0.440%

    No Known Activations