© 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. 15811
    Prev
    Next
    INDEX
    Explanations

    information related to scientific research on brain cells and their functions

    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
    640
    +0.29
    1.4%
    442
    +0.12
    0.6%
    235
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.29
    0.10
    442
    +0.12
    0.09
    160
    +0.10
    0.08
    Negative Logits
     estate
    -0.79
     wardrobe
    -0.79
     apost
    -0.76
     resign
    -0.74
     cabinet
    -0.73
     mustache
    -0.73
     appoint
    -0.72
     hust
    -0.72
     veto
    -0.72
     surrender
    -0.71
    POSITIVE LOGITS
    Researchers
    1.88
    Scientists
    1.77
    Explore
    1.49
    Using
    1.41
    Methods
    1.36
    Writing
    1.35
    ccording
    1.35
    Research
    1.33
    Studies
    1.30
    Study
    1.29
    Activations Density 0.238%

    No Known Activations