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

    Neuronpedia

    Jacobian LensNEW
    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. 17763
    Prev
    Next
    INDEX
    Explanations

    economic and political terms and concepts

    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
    314
    +0.19
    0.9%
    640
    +0.10
    0.5%
    490
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    152
    +0.19
    0.10
    314
    +0.10
    0.03
    102
    +0.10
    0.08
    Negative Logits
     embassies
    -0.78
     Clubs
    -0.76
     charms
    -0.76
     monuments
    -0.75
     Lanka
    -0.75
     magazines
    -0.74
     queues
    -0.74
     chiefs
    -0.74
     disturbances
    -0.72
     establishments
    -0.71
    POSITIVE LOGITS
    exclusive
    1.25
    sized
    1.25
    powered
    1.11
    issue
    1.11
    dimensional
    1.08
    themed
    1.07
    level
    1.06
    centric
    1.05
    approved
    1.03
    centered
    1.02
    Activations Density 0.196%

    No Known Activations