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. 11439
    Prev
    Next
    INDEX
    Explanations

    links to Google Drive files

    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
    247
    +0.11
    0.5%
    765
    +0.10
    0.5%
    373
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.11
    0.03
    373
    +0.10
    -0.01
    313
    +0.10
    0.03
    Negative Logits
    Column
    -0.64
     Advertisement
    -0.62
    ivas
    -0.62
    ONSORED
    -0.61
     STOP
    -0.60
    rontal
    -0.59
     Conclusion
    -0.59
    Lens
    -0.58
     Psychiat
    -0.58
    Tip
    -0.58
    POSITIVE LOGITS
    usercontent
    0.68
    cdn
    0.66
    plugin
    0.66
    google
    0.63
     Adobe
    0.63
     frameworks
    0.63
    izards
    0.62
    album
    0.62
    download
    0.62
    sd
    0.61
    Activations Density 0.057%

    No Known Activations