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 3072
    5. 8-RES_FS3072-JB
    6. 1355
    Prev
    Next
    INDEX
    Explanations

    references to specific brands, products, or locations

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    373
    +0.14
    0.7%
    39
    +0.10
    0.5%
    101
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    39
    +0.14
    0.07
    530
    +0.10
    0.06
    739
    +0.10
    0.04
    Negative Logits
    aunts
    -0.76
     manif
    -0.66
     flank
    -0.64
     manifold
    -0.62
    trap
    -0.62
     millionaires
    -0.61
    gui
    -0.61
    gap
    -0.59
    ploy
    -0.59
    asks
    -0.59
    POSITIVE LOGITS
     copyrighted
    1.06
     trademarks
    1.01
     copyright
    0.98
     affiliate
    0.94
     editorial
    0.89
    çīĪ
    0.87
    Disclaimer
    0.86
     Contribut
    0.82
     Copyright
    0.82
     Attribution
    0.82
    Activations Density 1.734%

    No Known Activations