© 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. 5413
    Prev
    Next
    INDEX
    Explanations

    organizations and entities associated with politics, government, and policy

    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
    220
    +0.11
    0.5%
    261
    +0.11
    0.5%
    120
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    220
    +0.11
    0.04
    36
    +0.11
    0.04
    362
    +0.10
    0.04
    Negative Logits
    icides
    -0.80
     torped
    -0.73
    enery
    -0.71
     flares
    -0.70
     dolphins
    -0.69
     selfies
    -0.69
    epad
    -0.68
    icidal
    -0.67
     crabs
    -0.66
     needles
    -0.66
    POSITIVE LOGITS
     nonprofit
    1.22
     watchdog
    1.11
     organization
    1.10
     organisation
    1.06
     consortium
    1.03
     acronym
    1.01
     umbrella
    1.00
     successor
    1.00
     nonpartisan
    0.99
     subsidiary
    0.98
    Activations Density 0.141%

    No Known Activations