© 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 EvalsExportsinterp-engineNEWAPI 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. 16936
    Prev
    Next
    INDEX
    Explanations

    terms related to medical conditions, especially those involving multiple sclerosis and other diseases

    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
    662
    +0.12
    0.5%
    99
    +0.11
    0.5%
    237
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    237
    +0.12
    0.02
    357
    +0.11
    0.02
    662
    +0.10
    0.02
    Negative Logits
    ESE
    -0.73
    iquette
    -0.72
    Dialogue
    -0.71
    NER
    -0.68
    OST
    -0.66
    ONEY
    -0.66
    kamp
    -0.65
    nothing
    -0.65
    anza
    -0.65
    naire
    -0.64
    POSITIVE LOGITS
     sclerosis
    1.63
    xes
    1.54
     iterations
    1.17
     layers
    1.09
     instances
    1.06
    ples
    1.05
     versions
    1.04
     facets
    1.02
     generations
    1.01
     tiers
    1.00
    Activations Density 0.032%

    No Known Activations