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. 1462
    Prev
    Next
    INDEX
    Explanations

    parts of text related to technology or digital platforms

    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
    217
    +0.14
    0.7%
    720
    +0.13
    0.6%
    498
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    720
    +0.14
    0.11
    217
    +0.13
    0.09
    498
    +0.11
    0.09
    Negative Logits
    SourceFile
    -0.64
     Coverage
    -0.62
    ORPG
    -0.62
    aukee
    -0.61
    igers
    -0.60
    olves
    -0.60
    ãĤ´ãĥ³
    -0.58
    crop
    -0.57
    izon
    -0.56
    FTWARE
    -0.55
    POSITIVE LOGITS
     alas
    1.18
     uh
    1.01
     unsurprisingly
    0.99
     um
    0.93
     somew
    0.92
     cru
    0.90
     interestingly
    0.89
     frankly
    0.88
     moreover
    0.87
     ironically
    0.85
    Activations Density 1.587%

    No Known Activations