© 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. 17292
    Prev
    Next
    INDEX
    Explanations

    tools, tips, and practices related to project management and collaboration

    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
    6
    +0.14
    0.7%
    640
    +0.13
    0.6%
    679
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.14
    0.09
    640
    +0.13
    0.07
    679
    +0.11
    0.07
    Negative Logits
     oun
    -0.85
    anmar
    -0.78
     satell
    -0.77
     notor
    -0.72
     charact
    -0.72
     confir
    -0.72
     glim
    -0.71
    bleacher
    -0.71
    berra
    -0.71
     rall
    -0.71
    POSITIVE LOGITS
     Also
    1.41
     Additionally
    1.23
     Specifically
    1.20
     Currently
    1.20
     Includes
    1.19
     Ideally
    1.19
     Typically
    1.19
     Previously
    1.18
     Instead
    1.18
     Usually
    1.15
    Activations Density 0.309%

    No Known Activations