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 49152
    5. 8-RES_FS49152-JB
    6. 45878
    Prev
    Next
    INDEX
    Explanations

    references to specific articles or products within the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    116
    +0.12
    0.5%
    393
    +0.11
    0.5%
    159
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    159
    +0.12
    0.04
    589
    +0.11
    0.03
    767
    +0.10
    0.03
    Negative Logits
    aus
    -0.81
    bats
    -0.72
    nings
    -0.69
    nown
    -0.69
    Ĭ±
    -0.68
     Izan
    -0.66
     Palest
    -0.64
     Mund
    -0.62
     Gavin
    -0.62
    ornings
    -0.61
    POSITIVE LOGITS
     article
    0.96
     item
    0.91
     ARTICLE
    0.87
     topic
    0.86
     slideshow
    0.85
     particular
    0.84
     repository
    0.82
     wiki
    0.82
     trope
    0.81
     addon
    0.79
    Activations Density 0.060%

    No Known Activations