Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 1511
    Prev
    Next
    INDEX
    Explanations

    percentages or numerical values related to economic data

    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
    507
    +0.11
    0.5%
    230
    +0.11
    0.5%
    10
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    260
    +0.11
    0.11
    507
    +0.11
    0.11
    230
    +0.10
    0.10
    Negative Logits
     redist
    -0.63
     retali
    -0.58
     grate
    -0.57
     cleanup
    -0.57
     renovations
    -0.57
     Dish
    -0.55
    wing
    -0.54
     regress
    -0.54
     spree
    -0.54
     reinvest
    -0.54
    POSITIVE LOGITS
    %.
    1.10
    %,
    1.00
    %;
    0.96
     apiece
    0.94
    fps
    0.83
    bps
    0.82
    0000000
    0.78
     Mbps
    0.77
    cm
    0.77
     degrees
    0.76
    Activations Density 0.952%

    No Known Activations