© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 10239
    Prev
    Next
    INDEX
    Explanations

    mentions of a specific newspaper, "The Wall Street Journal"

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    442
    +0.11
    0.5%
    70
    +0.11
    0.5%
    757
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    304
    +0.11
    0.02
    88
    +0.11
    0.02
    527
    +0.09
    0.01
    Negative Logits
    imum
    -0.81
    hered
    -0.74
    alogue
    -0.74
    ELY
    -0.74
    onna
    -0.73
    864
    -0.73
    netflix
    -0.72
    opl
    -0.68
    nah
    -0.67
    essa
    -0.67
    POSITIVE LOGITS
     Journal
    1.21
     firms
    1.09
     bankers
    0.99
     analysts
    0.94
     banker
    0.90
     Banking
    0.89
     banks
    0.88
     executives
    0.84
     analyst
    0.84
     finan
    0.83
    Activations Density 0.091%

    No Known Activations