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. 118
    Prev
    Next
    INDEX
    Explanations

    specific numerical information, such as quantities or rankings

    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
    21
    +0.12
    0.5%
    542
    +0.10
    0.5%
    157
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    21
    +0.12
    0.10
    5
    +0.10
    0.08
    616
    +0.10
    0.09
    Negative Logits
    utics
    -0.82
    each
    -0.72
    strength
    -0.71
    their
    -0.71
    allah
    -0.68
    rey
    -0.68
    lag
    -0.68
    terness
    -0.68
    erved
    -0.68
    apt
    -0.67
    POSITIVE LOGITS
     casualty
    1.17
     thing
    1.11
     reason
    0.94
     beneficiary
    0.93
     major
    0.91
     obstacle
    0.88
     culprit
    0.87
     installment
    0.86
     piece
    0.86
     exception
    0.85
    Activations Density 1.390%

    No Known Activations