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 24576
    5. 8-RES_FS24576-JB
    6. 23214
    Prev
    Next
    INDEX
    Explanations

    references to time periods such as weeks and months

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    430
    +0.14
    0.6%
    709
    +0.13
    0.6%
    30
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    430
    +0.14
    0.07
    310
    +0.13
    0.06
    30
    +0.12
    0.05
    Negative Logits
    EStreamFrame
    -0.71
    Condition
    -0.68
     grips
    -0.64
    ãĥĺ
    -0.63
    anium
    -0.63
    aeda
    -0.62
    ums
    -0.62
    unta
    -0.61
    -+
    -0.61
    hop
    -0.60
    POSITIVE LOGITS
     afternoon
    0.73
     meanwhile
    0.73
     HuffPost
    0.73
     rumors
    0.69
     reports
    0.68
     Ago
    0.67
     prosecutors
    0.66
     evening
    0.65
     Newsweek
    0.65
     Wikileaks
    0.65
    Activations Density 0.138%

    No Known Activations