© 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. 18321
    Prev
    Next
    INDEX
    Explanations

    names and titles related to political leaders and organizations

    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
    420
    +0.14
    0.7%
    481
    +0.13
    0.6%
    430
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    41
    +0.14
    0.01
    354
    +0.13
    0.01
    420
    +0.13
    0.01
    Negative Logits
    ESA
    -0.77
    CLA
    -0.72
    eries
    -0.71
    ACA
    -0.70
     DMV
    -0.70
    BAT
    -0.69
    pport
    -0.68
    ptroller
    -0.68
    EP
    -0.67
     Sussex
    -0.67
    POSITIVE LOGITS
    doms
    1.20
     Sheikh
    1.01
    istani
    0.99
     Hassan
    0.88
     Mohammed
    0.87
     Mohammad
    0.85
     Muhammad
    0.81
    astically
    0.81
     Ali
    0.81
     Allaah
    0.81
    Activations Density 0.024%

    No Known Activations