© 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. 10784
    Prev
    Next
    INDEX
    Explanations

    terms related to medical conditions and research studies, especially focusing on cancer and related treatments

    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
    1
    +0.11
    0.5%
    259
    +0.10
    0.4%
    265
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    252
    +0.11
    0.05
    655
    +0.10
    0.04
    546
    +0.09
    0.03
    Negative Logits
    soDeliveryDate
    -0.98
    Tokens
    -0.79
     desks
    -0.76
    atron
    -0.74
     rooft
    -0.73
     Bezos
    -0.72
     booth
    -0.72
     Vote
    -0.72
    displayText
    -0.71
    bucks
    -0.70
    POSITIVE LOGITS
     Symptoms
    1.50
     symptoms
    1.34
     disease
    1.28
     neurological
    1.25
     diseases
    1.24
     inflammation
    1.23
    ptoms
    1.23
     illnesses
    1.21
     syndrome
    1.18
     hereditary
    1.15
    Activations Density 0.421%

    No Known Activations