© 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. 24053
    Prev
    Next
    INDEX
    Explanations

    statements related to healthcare and its costs

    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
    481
    +0.14
    0.7%
    205
    +0.10
    0.5%
    194
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    8
    +0.14
    0.02
    384
    +0.10
    0.02
    512
    +0.10
    0.02
    Negative Logits
     Berk
    -0.82
     Refresh
    -0.71
     Benson
    -0.70
     unim
    -0.67
     Unch
    -0.67
     Boe
    -0.65
     Clayton
    -0.64
     Chu
    -0.64
    hare
    -0.63
     recl
    -0.63
    POSITIVE LOGITS
    about
    1.32
     about
    1.09
     ABOUT
    1.02
     About
    0.91
    About
    0.88
    abouts
    0.85
    deals
    0.77
    ²¾
    0.74
    hemat
    0.74
    poses
    0.73
    Activations Density 0.130%

    No Known Activations