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. 41257
    Prev
    Next
    INDEX
    Explanations

    mentions of time durations, specifically focusing on years and decades

    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
    571
    +0.12
    0.5%
    258
    +0.11
    0.5%
    377
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    657
    +0.12
    0.04
    651
    +0.11
    0.04
    258
    +0.10
    0.04
    Negative Logits
    edIn
    -0.65
    este
    -0.65
    ibrary
    -0.65
    ¬¼
    -0.62
     subp
    -0.61
    anooga
    -0.61
     Username
    -0.59
     Ft
    -0.59
    Ĥª
    -0.58
    Els
    -0.56
    POSITIVE LOGITS
     2020
    0.87
     2021
    0.86
    iband
    0.82
     someday
    0.81
    izons
    0.80
     2019
    0.80
     olds
    0.80
     hereafter
    0.79
    glass
    0.77
     unless
    0.75
    Activations Density 0.071%

    No Known Activations