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. 23652
    Prev
    Next
    INDEX
    Explanations

    numerical data and mentions of percentages

    oai_token-act-pair · gpt-4o-miniTriggered 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
    480
    +0.31
    1.6%
    87
    +0.19
    0.9%
    464
    +0.14
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    266
    +0.31
    0.08
    480
    +0.19
    0.03
    464
    +0.14
    0.08
    Negative Logits
    forth
    -0.70
    enegger
    -0.70
     Nieto
    -0.64
    coni
    -0.60
     beforehand
    -0.60
     resemb
    -0.57
     Seym
    -0.56
     Augustus
    -0.56
    /"
    -0.55
    terday
    -0.55
    POSITIVE LOGITS
     Featured
    0.80
    malink
    0.77
     guiActiveUnfocused
    0.72
     Location
    0.67
     Caption
    0.66
     Expand
    0.66
    enter
    0.66
     Submit
    0.65
     thumbnail
    0.65
     ·
    0.65
    Activations Density 0.636%

    No Known Activations