© 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. 22301
    Prev
    Next
    INDEX
    Explanations

    the repetition of the word "the" and other common words in a document

    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
    362
    +0.18
    0.9%
    481
    +0.16
    0.7%
    393
    +0.14
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.18
    0.06
    608
    +0.16
    0.05
    393
    +0.14
    0.04
    Negative Logits
    alion
    -0.73
    IFA
    -0.72
    udder
    -0.72
    owned
    -0.71
    ontent
    -0.71
    strate
    -0.70
    river
    -0.70
     accordingly
    -0.70
    ESA
    -0.69
    MRI
    -0.69
    POSITIVE LOGITS
     idea
    1.47
     notion
    1.13
     outcome
    1.07
     prospect
    1.06
     concept
    1.03
     possibility
    1.00
     opportunity
    1.00
     simplicity
    0.99
     variety
    0.92
     way
    0.91
    Activations Density 0.174%

    No Known Activations