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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 20202
    Prev
    Next
    INDEX
    Explanations

    words related to news articles and reporting

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    649
    +0.10
    0.5%
    496
    +0.10
    0.5%
    756
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    252
    +0.10
    0.02
    106
    +0.10
    0.02
    353
    +0.09
    0.02
    Negative Logits
    holes
    -0.95
    imately
    -0.88
    NetMessage
    -0.86
    lessly
    -0.80
    ¯
    -0.78
    boa
    -0.76
    docker
    -0.76
    artisan
    -0.75
    achu
    -0.75
    encers
    -0.73
    POSITIVE LOGITS
     Beast
    1.14
     Caller
    1.09
    motion
    0.99
     Mail
    0.98
     Inqu
    0.97
     Telegraph
    0.92
     Herald
    0.91
    Beast
    0.87
     Meal
    0.86
    Mail
    0.85
    Activations Density 0.018%

    No Known Activations