© 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. 20916
    Prev
    Next
    INDEX
    Explanations

    mentions of the term "Marine Corps."

    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
    288
    +0.14
    0.6%
    57
    +0.10
    0.5%
    200
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    152
    +0.14
    0.02
    288
    +0.10
    0.02
    200
    +0.10
    0.01
    Negative Logits
    reddits
    -0.83
    âĶģ
    -0.81
    rx
    -0.78
    ourcing
    -0.75
    ality
    -0.75
    otten
    -0.74
    Interstitial
    -0.72
    ible
    -0.72
    mble
    -0.71
    irable
    -0.71
    POSITIVE LOGITS
     Corps
    1.39
     Mamm
    1.05
     biologist
    1.01
     barracks
    0.97
     Biological
    0.89
     biologists
    0.86
     Conservation
    0.85
     colonel
    0.83
     Parade
    0.82
     corps
    0.82
    Activations Density 0.025%

    No Known Activations