© 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. 8698
    Prev
    Next
    INDEX
    Explanations

    mentions of medical professionals, specifically doctors and physicians

    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
    235
    +0.10
    0.5%
    212
    +0.10
    0.4%
    213
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    235
    +0.10
    0.03
    571
    +0.10
    0.02
    32
    +0.09
    0.02
    Negative Logits
     Territories
    -0.98
     Flavoring
    -0.76
     Rivals
    -0.64
     Passage
    -0.61
     Strongh
    -0.60
     Tactics
    -0.60
    izont
    -0.59
    wm
    -0.59
    ptions
    -0.59
     Dwar
    -0.59
    POSITIVE LOGITS
     prescribing
    0.94
     doctor
    0.93
     specializing
    0.90
    doctor
    0.86
     treating
    0.86
     diagn
    0.86
     diagnosed
    0.85
     diagnoses
    0.84
     prescribe
    0.84
     examiner
    0.84
    Activations Density 0.034%

    No Known Activations