© 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. 9010
    Prev
    Next
    INDEX
    Explanations

    references to superhero characters and related themes, particularly from the Marvel universe

    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
    450
    +0.10
    0.5%
    154
    +0.10
    0.4%
    160
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    450
    +0.10
    0.02
    154
    +0.10
    0.02
    425
    +0.09
    0.02
    Negative Logits
     embargo
    -0.70
     infl
    -0.70
     interstate
    -0.69
     redress
    -0.68
    assetsadobe
    -0.66
     chilled
    -0.65
     encour
    -0.65
     arrivals
    -0.64
     greeting
    -0.64
     interval
    -0.64
    POSITIVE LOGITS
    Woman
    1.02
    Cola
    0.99
    Robin
    0.98
    Ver
    0.94
    vill
    0.94
    Man
    0.89
    Sax
    0.89
    Girl
    0.88
    sense
    0.86
    Tro
    0.86
    Activations Density 0.036%

    No Known Activations