© 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. 24476
    Prev
    Next
    INDEX
    Explanations

    large quantities associated with complex issues, particularly related to social and political contexts

    oai_token-act-pair · gpt-4o-miniTriggered 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
    480
    +0.55
    3.0%
    87
    +0.19
    1.0%
    266
    +0.15
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.55
    0.04
    266
    +0.19
    0.12
    147
    +0.15
    0.05
    Negative Logits
     disadvant
    -0.55
    retty
    -0.53
    ensional
    -0.50
    vag
    -0.48
    ibrary
    -0.47
    rex
    -0.46
     Azerb
    -0.46
     Belg
    -0.45
    WARN
    -0.44
    jab
    -0.42
    POSITIVE LOGITS
     thereof
    0.94
     thereto
    0.90
     thereafter
    0.85
     therein
    0.73
    cellaneous
    0.72
     accordingly
    0.71
     alike
    0.62
     versa
    0.56
     afterward
    0.55
     accompanying
    0.54
    Activations Density 1.541%

    No Known Activations