© 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. 7729
    Prev
    Next
    INDEX
    Explanations

    medical terms, specifically related to chiropractic and vaccination controversies

    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
    +0.14
    0.7%
    107
    +0.11
    0.5%
    536
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    174
    +0.14
    0.03
    314
    +0.11
    0.01
    588
    +0.10
    0.03
    Negative Logits
     LINE
    -0.68
    blank
    -0.67
     Fault
    -0.65
     STE
    -0.62
    steen
    -0.62
     abundantly
    -0.61
     HIP
    -0.61
     Kepler
    -0.61
     SAP
    -0.61
     phased
    -0.61
    POSITIVE LOGITS
    orship
    1.04
    acist
    1.03
    ic
    0.98
    inical
    0.97
    icism
    0.96
    ics
    0.95
    ians
    0.95
    ners
    0.95
    ctors
    0.94
    ists
    0.94
    Activations Density 0.095%

    No Known Activations